Embed Size (px)
Citation preview
The relationship between lean service, activity-based costing
and business strategy and their impact on performance
A thesis submitted in partial fulfilment of the requirements for the
degree of Doctor of Philosophy
By
Wael Hadid
Brunel Business School
Brunel University London
April 2014
ii
Abstract
Lean system has drawn the attention of researchers and practitioners since its emergence in
1950s. This has been reflected by the increasing number of companies attempting to implement
its practices and the large number of researchers investigating its effectiveness and identifying
important contextual factors which affect its implementation. The rising level of interest in lean
system has led to the emergence of three distinctive streams of literature.
The first stream of literature has focused on the effectiveness of lean system. However, this
literature was limited as it mainly examined the additive impact of lean practices on operational
performance in the manufacturing context. The second stream of literature has focused on the
role the accounting system in the lean context. In this body of literature, there was an agreement
among researchers on the superiority of activity-based costing system (ABC) over the traditional
accounting system in supporting the implementation of lean practices. However, most studies in
this strand of literature were either conceptual or case-based studies. The third stream of
literature has focused on the fit between business strategy and lean system. However,
inconclusive results were reported in relation to the suitability of lean system to firms adopting
the differentiation strategy and others adopting the cost leadership strategy.
The aim of this study is to develop and empirically test a conceptual model which integrates the
three distinctive streams of literature to extend their focus and overcome their limitations. More
specifically, the model developed in the current study highlights not only the additive impact of
lean practices but also the possible synergy among those practices in improving both operational
and financial performance of service firms. In addition, the model brings to light the potential
intervening role of ABC in the strategy-lean association.
After identifying and reviewing the relevant literature, the socio-technical system theory and
contingency theory were used to develop the conceptual model and associated hypotheses. A
questionnaire instrument was designed to collect empirical data which was supplemented by
objective data from the Financial Analysis Made Easy database in order to empirically test the
conceptual model using partial least squares structural equation modelling (PLS-SEM).
The findings of this study indicated that while the technical practices of lean service improved
only the operational performance of service firms, the social practices enhanced both operational
iii
and financial performance. In addition, the two sets of practices positively interacted to improve
firm performance over and above the improvement achieved from each set separately. Moreover,
ABC was found to have a positive association with lean practice, and consequently an indirect
positive relation with firm operational performance. Finally, both the differentiation and cost
leadership strategy had a direct positive relationship with lean practices. However, while ABC
was found to partially mediate the differentiation-lean association, it suppressed the cost
leadership-lean association leading to a case of inconsistent mediation.
The current study contributes to the current literature at different levels. First, at the theoretical
level, this study develops a conceptual framework which crosses different streams of literatures
mainly, lean system literature, management accounting literature (with focus on ABC), and
business strategy literature. Unlike previous studies, by integrating the perspective of socio-
technical system theory and contingency theory, the model (i) highlights not only the additive
but also the synergistic effect of lean service practices on firm performance, (ii) brings to light
the direct impact of ABC and business strategy on lean service practices and the intervening role
of ABC due to which the business strategy is assumed to have also an indirect influence on lean
practices, and (iii) offers an alternative view on how ABC can improve firm performance by
enhancing other organisational capabilities (lean practices) which are expected to improve
performance . Second, at the methodological level, unlike previous studies, this study includes a
large number of lean service practices and contextual variables to report more precisely on the
lean-performance association. In addition, the inclusion of the financial performance dimension-
measured by secondary data- in the model besides the operational performance is critical to
understand the full capability of lean service in improving firm performance. Further, employing
a powerful statistical technique (PLS-SEM) provides more credibility to the results reported in
this study. Third, at the empirical level, this study is conducted in the UK service sector. As such,
this study is one of the very few studies that have reported on lean service and examined how the
adoption of ABC and a specific type of business strategy can affect its implementation using
empirical survey data from this context.
iv
Dedication
To my parents, brothers and sister for their enduring love, support and encouragement during my
journey to complete this thesis.
v
Acknowledgement
I would like to express my gratitude to my supervisors Dr. Afshin Mansouri and Dr. David
Gallear for their great help, support, advice, and encouragement throughout the PhD programme.
I would like to express my indebtedness to the Syrian Ministry of Higher Education for their
financial and academic support during my period of study.
Last but not least, all my thanks go also to my officemates, PhD colleagues, and friends with
whom I shared an important episode of my life.
vi
Declaration
I hereby declare that the materials contained in this thesis have not been previously submitted for
a degree in this or any other university. I further declare that this thesis is solely based on my
own research.
I also declare that all information in this document has been obtained and presented in
accordance with academic rules and ethical conduct.
Wael Hadid
vii
Publications associated with the thesis
Hadid, W. and Mansouri, A. (2014). The lean-performance relationship in services: A theoretical
model. International Journal of Operations and Production Management, 34(6), 750-785
viii
Table of Content
CHAPTER 1 : INTRODUCTION ............................................................................................... 1
1.1 Research background ................................................................................................................ 1
1.2 Research motivations ................................................................................................................ 5
1.3 Research problem...................................................................................................................... 7
1.4 Research questions .................................................................................................................... 9
1.5 Aim and objectives of the research ........................................................................................... 9
1.6 Research methodology ............................................................................................................ 10
1.7 Significance of the research .................................................................................................... 10
1.8 Structure of the thesis.............................................................................................................. 11
CHAPTER 2 : LITERATURE REVIEW ................................................................................ 14
2.1 Introduction ............................................................................................................................. 14
2.2 The evolution of the lean system ............................................................................................ 14
2.3 The structure and mechanism of lean system ......................................................................... 16
2.4 Empirical literature on the lean-performance association ...................................................... 19
2.4.1 The effect of isolated lean practices on firm performance ............................................... 20
2.4.2 The effect of lean bundles ................................................................................................ 22
2.4.3 The synergy between lean bundles................................................................................... 24
2.4.4 Critical evaluation of the empirical literature on the lean-performance association ........ 34
2.5 Lean system in the service sector............................................................................................ 35
2.5.1 The emergence of the lean service concept ...................................................................... 36
2.5.2 Lean service in the healthcare sector................................................................................ 38
2.5.3 Lean in other service sectors ............................................................................................ 41
2.5.3.1 Lean service and IT industry ..................................................................................... 42
2.5.3.2 Lean service and call centers ..................................................................................... 42
2.5.3.3 Lean service in the financial industry ........................................................................ 43
2.5.3.4 Lean service and telecommunication industry .......................................................... 44
2.5.3.5 Lean service and education ........................................................................................ 45
2.5.3.6 Lean service and housing services ............................................................................. 46
2.5.3.7 Lean service and the distribution industry ................................................................. 46
2.5.3.8 Lean service and office operations of manufacturing firms ...................................... 47
2.5.4 Evaluation of the lean service literature ........................................................................... 49
2.6 The literature on lean system, accounting system and business strategy ............................... 52
2.6.1 Accounting system and lean system................................................................................. 53
2.6.2 Business strategy and lean system ................................................................................... 56
2.6.3 Firm age and lean system ................................................................................................. 58
2.6.4 Firm size and lean system ................................................................................................ 58
2.6.5 Evaluation of the lean-context literature .......................................................................... 59
2.7 Conclusion and research gaps ................................................................................................. 60
CHAPTER 3 : THEORETICAL FRAMEWORK AND HYPOTHESES DEVELOPMENT
....................................................................................................................................................... 62
ix
3.1 Introduction ............................................................................................................................. 62
3.2 Constructs of the theoretical model ........................................................................................ 62
3.2.1 Identification of the relevant literature ............................................................................. 62
3.2.2 Identification of lean service practices ............................................................................. 63
3.2.3 The performance outcome of lean service practices ........................................................ 68
3.2.4 Contextual variables ......................................................................................................... 70
3.3 The theoretical foundation of the research.............................................................................. 70
3.3.1 Socio-technical systems theory (STS).............................................................................. 70
3.3.2 Contingency theory .......................................................................................................... 71
3.3.2.1 The selection approach .............................................................................................. 72
3.3.2.2 The interaction approach ........................................................................................... 73
3.3.2.3 The system approach ................................................................................................. 73
3.3.2.4 The moderation perspective ....................................................................................... 74
3.3.2.5 The mediation perspective ......................................................................................... 74
3.4 Theoretical model ................................................................................................................... 75
3.5 Research hypotheses ............................................................................................................... 76
3.5.1 The impact of lean service and contextual variables on firm performance ..................... 77
3.5.1.1 Lean technical practices (LTPs) and firm performance ............................................ 77
3.5.1.2 Lean social practices (LSPs) and firm performance .................................................. 79
3.5.1.3 The synergistic effect of LTPs and LSPs on performance ........................................ 80
3.5.2 The effect of contextual variables on performance and/or lean service ........................... 82
3.5.2.1 Firm age and firm performance ................................................................................. 82
3.5.2.2 Activity-based costing system and firm performance ............................................... 83
3.5.2.3 Firm size and firm performance ................................................................................ 86
3.5.2.4 Business strategy and lean technical practices .......................................................... 87
3.5.2.5 Business strategy and activity-based costing system ................................................ 89
3.5.2.6 Activity-based costing system and lean technical practices ...................................... 91
3.6 Summary of chapter 3 ............................................................................................................. 93
CHAPTER 4 : RESEARCH METHODOLOGY .................................................................... 96
4.1 Introduction ............................................................................................................................. 96
4.2 Research Paradigms ................................................................................................................ 96
4.2.1 Positivism ......................................................................................................................... 98
4.2.2 Interpretivism ................................................................................................................... 98
4.2.3 Rationale for adopting the positivist paradigm ................................................................ 99
4.3 Research approach (Deductive versus inductive approach) ................................................... 99
4.3.1 The deductive approach ................................................................................................. 100
4.3.2 The inductive approach .................................................................................................. 100
4.3.3 Rationale for adopting the deductive approach .............................................................. 102
4.4 Research strategy .................................................................................................................. 102
4.5 Data collection method ......................................................................................................... 104
4.5.1 Interview method............................................................................................................ 104
4.5.2 Questionnaire method .................................................................................................... 105
x
4.6 Research context ................................................................................................................... 106
4.7 Research population .............................................................................................................. 108
4.8 Research sampling ................................................................................................................ 108
4.9 Questionnaire development .................................................................................................. 110
4.9.1 Questions type and format ............................................................................................. 111
4.9.2 Questionnaire layout and questions flow ....................................................................... 111
4.9.3 Covering letter ................................................................................................................ 112
4.9.4 Questionnaire pre-testing process .................................................................................. 112
4.9.5 Questionnaire administration ......................................................................................... 113
4.9.5.1 Non-response bias.................................................................................................... 115
4.10 Variables measurement ....................................................................................................... 116
4.10.1 Lean service measurement ........................................................................................... 117
4.10.2 Organisational performance measurement ................................................................... 117
4.10.3 Contextual variables measurement............................................................................... 120
4.10.4 Common method bias................................................................................................... 120
4.11 Statistical techniques ........................................................................................................... 121
4.11.1 Structural Equation Modelling ..................................................................................... 121
4.11.2 Comparison between CB-SEM and PLS-SEM ............................................................ 122
4.11.3 Overview of PLS-SEM ................................................................................................ 123
4.11.3.1 Assessing the Measurement Model under PLS-SEM ............................................ 124
4.11.3.2 Assessing the Structural Model under PLS-SEM.................................................. 124
4.11.3.3 PLS-SEM Software ............................................................................................... 125
4.12 Summary of chapter four .................................................................................................... 125
CHAPTER 5 : DATA PREPARATION AND EXAMINATION ........................................ 127
5.1 Introduction ........................................................................................................................... 127
5.2 Preliminary screening of data ............................................................................................... 127
5.2.1 Missing data analysis ..................................................................................................... 127
5.2.2 Examining the assumptions of parametric tests ............................................................. 128
5.3 Validity, reliability and unidimensionality of constructs ...................................................... 129
5.3.1 Content validity .............................................................................................................. 129
5.3.2 Unidimensionality of constructs..................................................................................... 129
5.3.3 Reliability of constructs ................................................................................................. 130
5.3.3.1 The stability method ................................................................................................ 130
5.3.3.2 The internal reliability method ................................................................................ 131
5.3.4 Validity of constructs ..................................................................................................... 132
5.3.4.1 Convergent Validity ................................................................................................ 132
5.3.4.2 Discriminant Validity .............................................................................................. 132
5.3.5 Empirical examination of unidimensionality, reliability and validity of constructs ...... 133
5.3.5.1 Assessing unidimensionality, reliability and validity of LTPs ................................ 133
5.3.5.2 Assessing unidimensionality, reliability and validity of LSPs ................................ 134
xi
5.3.5.3 Assessing unidimensionality, reliability and validity of organisational performance
............................................................................................................................................. 135
5.3.5.4 Assessing unidimensionality, reliability and validity of business strategy ............. 136
5.4 Descriptive statistics ............................................................................................................. 137
5.4.1 Descriptive statistics of LTPs ......................................................................................... 137
5.4.2 Descriptive statistics of LSPs ......................................................................................... 139
5.4.3 Descriptive statistics of organisational performance ..................................................... 140
5.4.4 Descriptive statistics of business strategy ...................................................................... 141
5.4.5 Descriptive statistics of MAS ......................................................................................... 143
5.4.6 Descriptive statistics of internationalisation and Unionisation ...................................... 144
5.4.7 Descriptive statistics of firm age and size ...................................................................... 144
5.4.8 Correlations and multicollinearity .................................................................................. 145
5.5 Summary of chapter five ....................................................................................................... 147
CHAPTER 6 : STATISTICAL ANALYSIS USING PLS-SEM........................................... 148
6.1 Introduction ........................................................................................................................... 148
6.2 Measurement model .............................................................................................................. 148
6.2.1 Constructs of the measurement model ........................................................................... 148
6.2.2 Measurement model evaluation ..................................................................................... 149
6.2.2.1 Construct reliability ................................................................................................. 149
6.2.2.2 Construct validity (convergent and discriminant) ................................................... 150
6.3 Structural model .................................................................................................................... 152
6.3.1 Stages for hypotheses testing using PLS-SEM .............................................................. 154
6.3.2 Structural model specification and evaluation ............................................................... 156
6.4 Hypotheses testing ................................................................................................................ 157
6.4.1 Lean bundles, contextual variables and firm performance ............................................ 161
6.4.1.1 The main effect of lean technical bundles on firm performance ............................. 161
6.4.1.2 The main effect of lean social bundles on firm performance .................................. 161
6.4.1.3 The synergy between the social and technical bundles of lean service ................... 162
6.4.1.4 Firm age and performance ....................................................................................... 162
6.4.1.5 ABC and performance ............................................................................................. 163
6.4.1.6 Firm size and performance ...................................................................................... 163
6.4.2 Lean technical bundles, ABC and business strategy ...................................................... 163
6.4.2.1 Differentiation strategy and lean technical bundles ................................................ 163
6.4.2.2 Cost leadership strategy and lean technical bundles ............................................... 164
6.4.2.3 Differentiation strategy and ABC ............................................................................ 164
6.4.2.4 Cost leadership strategy and ABC ........................................................................... 164
6.4.2.5 ABC and the technical bundles of lean service ....................................................... 164
6.5 Summary of chapter six ........................................................................................................ 166
CHAPTER 7 : DISCUSSION AND IMPLICATIONS ......................................................... 167
7.1 Introduction ........................................................................................................................... 167
7.2 Lean service bundles, contextual variables and firm performance ....................................... 167
xii
7.2.1 The direct relationship between LTP factors and firm performance ............................. 167
7.2.2 The direct relationship between LSP factors and firm performance .............................. 170
7.2.3 The synergistic effect of LSPs and LTPs on firm performance ..................................... 171
7.2.4 Firm age and firm performance ...................................................................................... 172
7.2.5 ABC and firm performance ............................................................................................ 172
7.2.6 Firm size and firm performance ..................................................................................... 173
7.3 Lean service, ABC and business strategy ............................................................................. 174
7.3.1 LTPs and Business strategy ........................................................................................... 174
7.3.2 Business strategy and ABC ............................................................................................ 175
7.3.3 LTPs and ABC ............................................................................................................... 176
7.3.4 ABC, business strategy and LTPs .................................................................................. 177
7.4. Summary of chapter seven ................................................................................................... 178
CHAPTER 8 : CONCLUSIONS ............................................................................................. 180
8.1 Introduction ........................................................................................................................... 180
8.2 Summary of research findings .............................................................................................. 181
8.3 Revisiting the research questions .......................................................................................... 182
8.3.1 The first research question ............................................................................................. 182
8.3.2 The second research question ......................................................................................... 183
8.4 Research Contributions ......................................................................................................... 184
8.5 Limitations and Directions for Future Research ................................................................... 185
REFERENCES .......................................................................................................................... 188
APPENDIX (1): THE QUESTIONNAIRE ............................................................................ 223
APPENDIX (2): INTRODUCTORY LETTER ..................................................................... 227
APPENDIX (3): GLOSSARY SHEET ................................................................................... 228
xiii
List of Tables
Table 2-1: Summary of the lean-performance association literature ............................................ 27
Table 3-1: Lean technical practices ............................................................................................. 65
Table 3-2: Lean social practices ................................................................................................... 67
Table 3-3: Benefits of lean service ............................................................................................... 68
Table 3-4: Summary of the research hypotheses .......................................................................... 95
Table 4-1: The assumptions of the positivism and interpretivism paradigms .............................. 97
Table 4-2: The main features of the positivism and interpretivism paradigms ............................ 98
Table 4-3: The main differences between the deductive and inductive approaches................... 101
Table 4-4: Research strategies under the two main research paradigms .................................... 102
Table 4-5: Reasons for non-participation ................................................................................... 115
Table 4-6: The sample distribution ............................................................................................. 115
Table 5-1: Factor analysis and reliability analysis of lean technical practices ........................... 134
Table 5-2: Factor analysis and reliability analysis of lean social practices ................................ 135
Table 5-3: Factor analysis and reliability analysis of operational performance ......................... 136
Table 5-4: Factor analysis and reliability analysis for cost leader strategy ................................ 138
Table 5-5: Factor analysis and reliability analysis for differentiation strategy .......................... 138
Table 5-6: Descriptive statistics of LTPs factors and indicators ................................................ 139
Table 5-7: Descriptive statistics of LSP factors and indicators .................................................. 140
Table 5-8: Descriptive statistics of operational performance factors and indicators .................. 142
Table 5-9: Descriptive statistics of financial performance ......................................................... 142
Table 5-10: Descriptive statistics of business strategy ............................................................... 143
Table 5-11: Descriptive statistics of MAS .................................................................................. 143
Table 5-12: Descriptive statistics of internationalisation and unionisation ................................ 144
Table 5-13: Descriptive statistics of firm age and firm size ....................................................... 145
Table 5-14: The correlation matrix of the independent variables ............................................... 146
Table 5-15: The correlation matrix of all DVs and IVs of this study ......................................... 146
Table 6-1: Reliability and validity measures of constructs ......................................................... 151
Table 6-2: The correlation matrix of constructs and square root of AVE .................................. 152
Table 6-3: Correlation matrix of constructs and indicators ........................................................ 153
Table 6-4: Constructs included in the structural model with respective reasons for their inclusion
............................................................................................................................................. 154
Table 6-5: Cross validated Communality and Redundancy ....................................................... 157
Table 6-6: Lean bundles, contextual variables and firm performance-results from PLS-SEM.. 158
Table 6-7: Lean bundles interaction and firm performance- results from PLS-SEM................. 159
Table 6-8: Summary table of hypotheses testing ........................................................................ 165
xiv
List of figures
Figure 2-1: The classification of lean service literature per industry type ................................... 50
Figure 2-2: The trend of lean service publications ....................................................................... 51
Figure 3-1: The moderation perspective ....................................................................................... 74
Figure 3-2: The mediation perspective ......................................................................................... 75
Figure 3-3: The theoretical model................................................................................................. 77
Figure 3-4: The mechanism of the socio-technical system theory ............................................... 81
Figure 4-1: The six stages of the deductive approach ................................................................. 100
Figure 4-2: Summary of the methodological choices adopted in this research .......................... 107
Figure 6-1: The measurement and structural model ................................................................... 150
xv
List of abbreviations
ABC: Activity-based costing
AVE: Average variance extracted
AMT: Advanced manufacturing technology
CT: Contingency theory
CFL: Changing facility layout
CFA: Confirmatory factor analysis
DV: Dependent variable
EFA: Exploratory factor analysis
HRM: Human resource management
IV: Independent variable
JIT: Just in time
LTPs: Lean technical practices
LSPs: Lean social practices
MAS: Management accounting system
MCS: Management control systems
NVAs: Non value adding activities
PM: Profit margin
PLS: Partial least squares
ROI: Return on investment
ROCE: Return on capital employed
STS: Socio-technical system theory
SEM: Structural equation modeling
TAS: Traditional costing systems
TE: Turnover per employee
TQM: Total quality management
TPM: Total preventive maintenance
VSM: Value stream mapping
1
Chapter 1 : Introduction
1.1 Research background
Globalisation and increasing competition have changed the market environment in which
companies across different sectors operate (Alsmadi et al., 2012; Soltani et al., 2010;
Askarany et al., 2010; Karmarkar, 2004). To survive in the new market environment,
companies should tackle several conflicting aspects amongst which are increasing
customers’ expectations, achieving higher revenues and increasing expenses (Askarany et
al., 2010; Allway and Corbett, 2002; Clarke and Mullins, 2001). To worsen this situation,
increasing prices of products/services in order to achieve a pre-specified revenue is no
longer an option for many companies across various industries (Stuebs and Sun, 2010).
Consequently, the ancient formula (i.e. selling price = cost + profit margin) that has
long been used to determine the selling price of products/services by adding a profit
margin to the cost of products/services is not as valid as it has been in the past (Yu-Lee,
2011). As a result, it seems more effective for most firms in this new era to focus on and
control their costs as a means for improving the increasingly uncontrollable revenue
(Askarany et al., 2010; Stuebs and Sun, 2010; Clarke and Mullins 2001). To achieve that,
new methods that are designed to analyse and control operations cost and satisfy
customers’ expectations should be adopted (Vlachos and Bogdanovic, 2013; Karmarkar,
2004).
The lean system has been proposed as an ideal, strategic option to confront the conflicting
aspects facing companies in the new market environment (Arlbjørn and Freytag, 2013).
Shah and Ward (2007, P.791) have defined lean system as “an integrated socio-technical
system whose main objective is to eliminate waste by concurrently reducing or
minimising supplier, customer, and internal variability”. The elimination of activities that
customers do not value is likely to lead to cost reductions, revenue improvements and
higher customer satisfaction (Vlachos and Bogdanovic, 2013; Womack and Jones, 1996).
Due to these expected benefits, lean practices have received special attention in practice
2
and academia (Taylor and Taylor, 2009; Maskell and Kennedy, 2007; Atkinson, 2004).
This has been reflected by the growing number of researchers who have attempted to
empirically validate the effectiveness of lean practices in delivering their purported
benefits to adopters (e.g. Agarwal et al., 2013; Bonavia and Marin-Garcia, 2011; Rahman
et al., 2010; De Menezes and Wood, 2006; Shah and Ward, 2003).
However, a common theme among most studies contributing to this stream of literature is
(a) their focus on manufacturing operations (Arlbjørn and Freytag, 2013; Malmbrandt and
Åhlström, 2013; Vlachos and Bogdanovic, 2013), (b) investigating mainly the additive
impact (i.e. the independent effect of practices) of lean practices on operational
performance with inconclusive results (Dabhilkar and Åhlström, 2013; Das and Jayaram,
2007), and (c) their general neglect to the potential influence of adopters’ organisational
context on the adoption of lean practices (Arlbjørn and Freytag, 2013; Fullerton et al.,
2013; Malmbrandt and Åhlström, 2013; Shah and Ward, 2003). More specifically, the role
of management accounting system (MAS) and business strategy in the adoption of lean
practices have not been rigorously investigated. These shortcomings limit our knowledge
on several aspects of the lean system and its integration with the context in which it
operates in a number of ways.
First, there is limited empirical evidence on the possible non-additive (i.e. the interaction
between practices) impact of lean practices on firm performance especially on financial
indicators despite their importance to top management (Arlbjørn and Freytag, 2013;
Camacho-Miñano et al., 2013; Dabhilkar and Åhlström, 2013; Das and Jayaram, 2007).
This limited focus to only the additive impact of lean practices prevents the realisation of
the full potential of lean system and the mechanism through which it affects firm
performance. This, in turn, may lead to hindering the spread of lean system among firms
and industries more widely (Fullerton and Wempe, 2009). Recently, a few researchers
have highlighted the need to adopt the socio-technical system theory (STS) when
examining the lean-performance association (e.g. Dabhilkar and Åhlström, 2013; Furlan et
al., 2011; Das and Jayaram, 2007). The STS views organisations to be consisted of two
separate, but interdependent, systems: a technical system and a social system (Appelbaum,
1997; Fox, 1995; Trist, 1981). The technical system comprises equipments, tools,
3
techniques and processes, while the social system comprises people and relationships
among them (Trist, 1981; Trist and Bamforth, 1951). The argument in this theory is that
optimal performance can be obtained only when there is a simultaneous implementation of
practices from both the technical and social system (Fox, 1995; Huber and Brown, 1991;
Trist and Bamforth, 1951). Therefore, although each system can be described as a stand-
alone system with its own benefits, the optimal performance of an organisation can only
be obtained by the joint optimisation of both systems (Zu, 2009; Manz and Stewart, 1997;
Fox, 1995; Trist, 1981; Emery and Trist, 1965). Such argument implies that it is possible
that the two systems of the STS interact to produce the highest possible outcome. To date,
little work has been done on perceiving lean system as a socio-technical system and
examining the interaction between the technical and social systems in improving firm
performance ( Dabhilkar and Åhlström, 2013; Furlan et al., 2011; Das and Jayaram,
2007).
Second, in spite of claims by proponents of lean service on the applicability and
effectiveness of lean practices in the service context, very scant empirical information has
been reported to date on the additive and non-additive effect of lean practices on the
performance of service firms (Arlbjørn and Freytag, 2013; Malmbrandt and Åhlström,
2013; Vlachos and Bogdanovic, 2013). In its current status, the literature on lean system in
services cannot be relied on to make informed conclusions on whether the applicability of
lean system is universal across sectors or only relevant to manufacturing operations where
it originated (Arlbjørn and Freytag, 2013; Malmbrandt and Åhlström, 2013).
Third, there is also a lack of sufficient information on what and how contextual factors can
impact the adoption of lean practices (Arlbjørn and Freytag, 2013; Fullerton et al., 2013;
Malmbrandt and Åhlström, 2013; Shah and Ward, 2003). From the contingency theory
(CT) perspective, any management or organisational system cannot be similarly effective
in all contexts (Rashidirad et al., 2013; Drazin and Van de Ven, 1985). As a result, a
specific context can be more conducive for a specific system than other contexts, which
positions the concept of fit at the heart of CT (Drazin and Van de Ven, 1985).
4
To date, there have been two streams of literature. The first stream of literature has
attempted to highlight the role of MAS in the implementation of lean practices (e.g.
Chiarini, 2012; Li et al., 2012; Cooper and Maskell, 2008; Kennedy and Widener, 2008;
Maskell and Kennedy, 2007; Maskell, 2006; Grasso, 2005; Carnes and Hedin, 2005;
Karlsson and Ahlstrom, 1996; South, 1993; Datar et al., 1991). The MAS focuses on
collecting, analysing, and then providing managements with accurate information on
resources consumed by their operations which can be used for different purposes
including stock valuation, pricing, make/buy decisions, operating performance evaluation,
planning and supporting improvement initiatives (Maiga and Jacobs, 2008; Fullerton and
McWatters, 2004). The core argument in this stream of literature is that traditional
accounting system (TAS) may hinder the adoption of lean practices and a more
compatible system such as activity-based costing system (ABC) (e.g. Khataie and Bulgak,
2013; Chiarini, 2012) or lean accounting (Ruiz-de-Arbulo-Lopez et al., 2013; Cooper and
Maskell, 2008) should be relied upon to support such adoption. However, very little
empirical research exists which has specifically examined the role of ABC and/or lean
accounting in the lean context (Fullerton et al., 2013; Banker et al., 2008). While
conducting empirical research on lean accounting may be premature as it is a relatively
recent concept and some companies may not be familiar with it (Ruiz-de-Arbulo-Lopez et
al., 2013; Chiarini, 2012), ABC has been around for over two decades and a considerable
number of companies have implemented it (Askarany et al., 2010; Kallunki and Silvola,
2008; Gosselin, 1997; Cooper and Kaplan, 1992).
The second stream of literature has empirically investigated the possible impact of
business strategy on lean practices (e.g. Qi et al., 2011; Ward et al., 2007; Chenhall and
Langfield-Smith, 1998). Business strategy usually expresses how a company chooses to
compete against its competitors and attain a competitive advantage in the market
(Bruggeman and Stede, 1993; Porter, 1980). Adopting the typology proposed by Porter
(1980) which classified companies into differentiators or cost leaders, studies in this
stream of literature have reported mixed results in terms of the relationship between lean
system and business strategy. While Qi et al. (2011) have found cost leadership strategy to
be more compatible with lean system than the differentiation strategy; others have
5
demonstrated the reverse (Ward et al., 2007; Chenhall and Langfield-Smith, 1998).
Studies in the two streams of literature presented above have a narrow focus on either the
accounting-lean relationship (e.g. Fullerton et al., 2013; Banker et al., 2008) or the
strategy-lean association (e.g. Qi et al., 2011; Ward et al., 2007). Consequently, neither of
the reviewed studies has adopted a wider view by simultaneously examining the impact of
the accounting system and business strategy on the lean system in one model. This
simultaneous examination is critical for two reasons. First, it allows for highlighting the
possible intervening role of the accounting system in the strategy-lean association. The
intervening role of the accounting system can be uncovered by integrating the findings of
the two previously reviewed streams of literature with the findings of studies documenting
the impact of business strategy on the accounting system (e.g. Hammad et al., 2010;
Auzair and Langfield-Smith, 2005; Chenhall, 2003; Gosselin, 1997).
Second, due to the anticipated impact of business strategy on both the accounting system
and the lean system (Qi et al., 2011; Chenhall and Langfield-Smith, 1998; Chenhall, 2003;
Gosselin, 1997), studying the impact of the accounting system on the lean system without
accounting for the effect of business strategy will provide biased findings due to
endogeneity issues caused by correlated omitted variables1 (Larcker and Rusticus, 2007).
Although the potential for endogeneity is present in almost all studies (Chenhall and
Moers, 2007), including the impact of business strategy when studying the lean-
accounting association addresses, at least partially, this issue.
1.2 Research motivations
This research is motivated by (1) the popularity of lean system to improve performance,
(2) the importance of the service sector to most developed economies, (3) the urgent need
to uncover and understand the full mechanism of lean system through which it impacts
firm performance, (4) the importance of clarifying the full role of the accounting system
1 Endogeneity caused by correlated omitted variables occurs as a result of excluding one or more variables which are expected to
affect both the dependent and independent variable(s) in a model (Chenhall and Moers, 2007).
6
and business strategy in the implementation of lean practices, and (5) calls from several
researchers for empirical studies to overcome the aforementioned limitations.
As mentioned before, the lean system has received a great deal of attention in both
practice and academia due to the purported benefits that can be attained from its adoption
(Taylor and Taylor, 2009; Maskell and Kennedy, 2007; Atkinson, 2004). Maskell and
Kennedy (2007) report that around 50% of American manufacturing companies strive to
achieve some level of a lean system in their plants. In addition, Taylor and Taylor (2009)
find that lean methods are one of the eight topics that have received most of the attention
in operations management research.
Furthermore, the contribution of the service sector to the gross domestic product
outweighs that of other sectors in most developed economies (Malmbrandt and Åhlström,
2013; Soltani et al., 2012; Chase and Apte, 2007; Chenhall, 2003). Therefore, conducting
more research to verify anecdotal evidence on the applicability and the effectiveness of
lean practices in the service sector has the potential to provide significant benefits, perhaps
at economy level (Apte and Goh, 2004; Brignall, 1997). If lean practices are proved to be
effective, then more service companies must be encouraged to adopt them and enjoy their
benefits (Fullerton and Wempe, 2009). On the other hand, if they are proved to be
ineffective, the potential adopters can stop experimenting with those practices, saving
time, effort and the resources required for their implementation (Arlbjørn and Freytag,
2013; Pavnaskar et al., 2003).
In addition, this research is also motivated by the serious limitations of the current lean
system literature that has not provided insightful information on different aspects of lean
system. First, as previously highlighted, there is a strong need to probe deeper into the
mechanism through which lean system impacts firm performance. This necessitates going
beyond the traditional lean-performance model by investigating not only the additive, but
also the non-additive effect of lean practices on firm performance (Dabhilkar and
Åhlström, 2013; Das and Jayaram, 2007; Shah and Ward, 2003). Second, the current lean
literature falls short in terms of providing sufficient evidence on the capability of lean
7
practices to improve the financial performance of adopters although the financial
performance is critical to the top management of most companies (Arlbjørn and Freytag,
2013; Camacho-Miñano et al., 2013; Womack and Jones, 1994). Overcoming both or
either of these limitations is expected to allow for a better understanding of the full
capability of lean practices. Third, investigating these two aspects of lean system in the
service rather than manufacturing context cannot be overestimated (Malmbrandt and
Åhlström, 2013; Suárez-Barraza et al., 2012; Voss, 2005). By doing so, we are able to
examine the universality of lean system at sector level. In other words, examining lean
system in the service context will provide evidence whether the manufacturing-oriented
system can be effective in different contexts or it is limited to its origin. Fourth, more
insights can be obtained on the role of MAS and business strategy in the lean context by
integrating three distinctive streams of literature, namely the lean-accounting literature, the
strategy-lean literature and the strategy-accounting literature. This integration allows for
developing a more complex model which, through the mediation perspective of the
contingency theory, can highlight the intervening role of MAS in the strategy-lean
association and decompose the strategy effect into its direct and indirect elements (Luft
and Shields, 2003; Shields et al, 2000). Consequently, an empirical examination of the
new model provides a better understanding of the role played by both the MAS and
business strategy in the adoption of lean practices which has been called for by several
researchers (e.g. Fullerton et al., 2013; Malmbrandt and Åhlström, 2013).
1.3 Research problem
The research problem stems from the desire and urgent need to go beyond the traditional,
limited lean-performance model and lean-context model that have been adopted by most
researchers to date. The traditional lean-performance model focuses mainly on whether
lean practices have any additive effect on operational performance of manufacturing firms.
As such, this model does not take into account any likely collaborative influence of lean
practices on performance (Dabhilkar and Åhlström, 2013; Kim et al., 2012; Shah and
Ward, 2003). Investigating synergy resulting from possible interaction between lean
practices could have a profound effect on how to best allocate resources among lean
practices. In addition, providing rigorous findings on the outcome of lean practices at
8
financial level can be as important as reporting on operational performance, or even more
so (Camacho-Miñano et al., 2013; Womack and Jones, 1994).
Allway and Corbett (2002) report that the lean practices used by service firms are very
similar to those already adopted by manufacturing firms. However, there is no conclusive
evidence on whether these practices originated in manufacturing can be effective in
services as they have been claimed to be in manufacturing (Arlbjørn and Freytag, 2013;
Malmbrandt and Åhlström, 2013; Vlachos and Bogdanovic, 2013). Our knowledge on
lean service to date has mainly been derived from conceptual and case studies literature
(Suárez-Barraza et al., 2012; Holm and Ahlstrom, 2010b). These types of research cannot,
due to their known inherent limitations, provide practical and generalisable evidence on
the effectiveness of lean service (Challis et al., 2002; Voss et al., 2002; Shields, 1995). To
worsen this situation, anecdotal evidence supporting the effectiveness of lean service has
faced some contrasting empirical results indicating that not all practices implemented in
manufacturing could be relevant or effective in services (Alsamdi et al., 2012; Yasin et al.,
2003). These findings are not unexpected given the well documented differences between
the two sectors (Soltani et al., 2012; Sampson and Froehle, 2006; Nie and Kellogg, 1999).
Added to that, the sole focus by the vast majority of researchers on verifying the lean-
performance association has directed attention away from the importance of understanding
the role of MAS and business strategy in the lean context (Fullerton et al., 2013;
Malmbrandt and Åhlström, 2013; Shah and Ward, 2003). However, even when the role of
MAS and business strategy has been brought to light by some researchers, these variables
have been examined in isolation (e.g. Fullerton et al., 2013; Qi et al., 2011; Amoako-
Gyampah and Acquaah, 2008; Banker et al., 2008; Ward et al., 2007; Fullerton and
McWatters, 2002). With this limited focus, only the additive effect of MAS and business
strategy has been highlighted. This conceals more insightful findings that can be obtained
from models which adopt a deeper view through decomposing the additive effect into its
direct and indirect elements (Luft and Shields, 2003; Shields et al., 2000).
9
Collectively, these shortcomings represent the research problem that will be addressed in
this thesis.
1.4 Research questions
To address the research problem discussed above, this study seeks to answer the following
questions:
1. Do lean practices have an additive and/or non-additive impact on operational
and/or financial performance of service firms?
2. Does each of the two contextual variables MAS and business strategy affect lean
service practices? If so, is the effect of each variable independent of the other
variable?
Several hypotheses are developed and reported later in chapter 3 to facilitate answering
these research questions.
1.5 Aim and objectives of the research
The aim of this research is to go beyond the traditional lean-performance model and lean-
context model by (a) focusing on not only the additive impact of lean practices on the
performance of service firms, but also on their likely interaction impact, and (b)
decomposing the additive impact of MAS and Business strategy on the adoption of lean
practices into its direct and indirect elements.
To facilitate the achievement of the research aim, the following objectives are set out:
1) To conduct a systematic review of the relevant literature in order to identify lean
service practices, performance indicators and contextual variables.
2) To develop a theoretical model that brings to light the mechanism through which
the identified variables are linked.
3) To empirically assess relationships hypothesised in the theoretical model in the
service context.
10
1.6 Research methodology
To achieve the aim and objectives of this study, a systematic search for lean service
publications was conducted by surveying publications in five well-known databases using
key words including “lean”, “process improvement”, “system thinking” and “more with
less”. Those databases include: Business Source Premier, ABI/INFORM Research,
Emerald, Science Direct and Scopus. In addition, a cross-sectional survey methodology
relying on a questionnaire instrument developed for the current research was adopted. This
is important for reporting more generalisable findings on the full potential of lean system
and the impact of MAS and business strategy on the adoption level of the system. The
questionnaire targeted mainly the operations managers of a sample of UK service firms
identified via the Financial Analysis Made Easy database (FAME). The empirical analysis
was conducted by utilising two appropriate and effective statistical techniques, namely
Factor Analysis and Partial Least Squares Structural Equation Modelling (PLS-SEM).
1.7 Significance of the research
This study offers valuable insights to different streams of literature especially the lean
service literature and the management accounting literature. This study provides fresh
evidence from a large-scale survey on both the additive and non-additive impact of lean
practices on the performance of service firms. Therefore, it is one of the very few studies
that have adopted the survey methodology to examine lean practices in services (Arlbjørn
and Freytag, 2013; Malmbrandt and Åhlström, 2013; Suárez-Barraza et al., 2012). The
findings of this study are then critical as they help to determine whether lean system is
universalistic across sectors or context-specific (Malmbrandt and Åhlström, 2013).
Moreover, the focus on both additive and non-additive impact of lean practices allows for
better understanding of the mechanism through which lean practices influence
performance (Dabhilkar and Åhlström, 2013; Shah and Ward, 2003). Consequently,
services managers can make informed decisions on how to best allocate resources among
those practices. In addition, surveying various service industries undoubtedly generates
more generalisable evidence (Wagner et al., 2012; Kallunki and Silvola, 2008; Challis et
al., 2002).
11
Investigating the impact of MAS, represented by the activity-based costing system (ABC),
and business strategy on the adoption of lean practices in services will also contribute
significantly to the current literature in different ways. First, this study highlights the
intervening role of ABC in the strategy-lean association. Therefore, the role of ABC as
both a dependent and independent variable is simultaneously examined in the same model,
in contrast to most accounting studies that treated the accounting system as either a
dependent or independent variable (Fullerton et al., 2013; Banker et al., 2008; Luft and
Shields, 2003). Second, most studies in the ABC literature have focused on the direct
effect of ABC implementation on performance with inconclusive results (e.g. Sheu and
Pan, 2009; Cagwin and Bouwman, 2002; Ittner et al., 2002; Kennedy and Affleck-Graves,
2001; Mishra and Vaysman, 2001; Shields, 1995). The current study provides an
alternative view on how ABC may indirectly improve performance by focusing on its
ability to develop other organisational capabilities (i.e. lean service) which, in turn, are
expected to improve firm performance. To date, this alternative perspective has been
rarely empirically investigated (Banker et al., 2008; Ittner et al., 2002). Finally, providing
empirical evidence on the effect of ABC and business strategy on lean service helps
service managers who are contemplating the implementation of lean service to be
cognizant of that effect on lean service.
1.8 Structure of the thesis
This chapter has provided a brief background of the study, explicitly presented the
research motivations, established the research problem and questions, and conveyed the
aim and objectives along with the significance of this research. The remainder of this
thesis is structured as follows:
Chapter 2 presents a review of the literature relevant to this current research and highlights
the research gaps. More specifically, this chapter will provide some brief information on
the evolution of lean system since its emergence in 1950s. This includes a review of the
various definitions reported by different researchers along with highlighting the confusion
surrounding the concept of lean system. In addition, the structure and mechanism through
12
which lean system is argued to improve the operations and performance of adopters are
presented. Finally, the empirical literature of lean system and the literature of lean service
are reviewed separately and critically assessed to expose their knowledge limitations and
propose avenues for overcoming the determined limitations. The last section of this
chapter details the shortcomings of the current lean literature that are addressed by the
current study.
Chapter 3 presents the theoretical model and associated hypotheses aimed to be examined
in this study. In this chapter the model constructs are identified. In addition, the theoretical
foundation of the developed model of this study is discussed. More specifically, the notion
of the socio-technical theory and the contingency theory are explicated and the importance
of their use in the current study is highlighted. Finally, the research theoretical model is
developed along with a number of hypotheses to be empirically examined in order to
satisfy the aim and objectives of this research study.
Chapter 4 outlines the research methodology adopted in the current study. More
specifically, explanation of the two main research paradigms (positivism versus
interpretivism) is provided along with the rationale behind the adoption of the positivism
paradigm. In addition, discussion of the different research approaches (deductive versus
inductive) and research strategies is presented accompanied by justification of the choices
made in adopting the deductive approach and the cross-sectional survey strategy.
Moreover, comparison of the different data collection methods is provided and the various
stages of developing the questionnaire instrument are reported. This chapter also identifies
the research context, population and the sample from which the data will be collected.
This is accompanied by detailed description of the study variables measured and process
of administering the questionnaire instrument. Finally, a description of the statistical
techniques along with the rationale behind their adoption will be also provided.
Chapter 5 elucidates the procedures employed to clean the empirical data collected and
examine it against the assumptions of parametric tests. This includes a detailed discussion
of the missing data analysis and the known assumption of parametric tests including
13
normality, linearity, and homogeneity of variance, outliers and multicollinearity. In
addition, this chapter establishes the unidimensionality of constructs and examine their
reliability and validity. Finally, descriptive statistics of all constructs are presented.
Chapter 6 illustrates the development and assessment of the PLS-SEM measurement and
structural model in the second and third sections, respectively. The fourth section of this
chapter presents the results of hypotheses testing and the last section provides a summary
of this chapter.
Chapter 7 presents a detailed discussion of the hypotheses testing results reported in
chapter 6. This chapter attempts to position the results achieved for each of the research
hypotheses within the relevant extant literature so that differences are highlighted and
implications are deduced.
Chapter 8 provides a summary of this research and report the main conclusions based on
findings from testing the research hypotheses. This chapter also highlights the limitations
of the current study and provides avenues for future research.
14
Chapter 2 : Literature review
2.1 Introduction
The lean system has attracted the attention of many academics and practitioners over the
last few decades (Taylor and Taylor, 2009; Maskell and Kennedy, 2007; Atkinson, 2004).
However, what do we mean by lean system? What is the structure of a lean system? What
is the mechanism through which the lean system improves adopters’ performance? What
limitations does the current lean literature suffer from which hinders better insights to be
obtained? This chapter revolves around providing answers to such questions. To achieve
this aim, the chapter is divided into seven sections. The second section discusses the
evolution of lean system highlighting its various definitions and the confusion surrounding
the concept. The third section is devoted to clarify the structure of lean system and the
mechanism through which it is expected to influence firm performance. Section four is
dedicated to present and critically assess the empirical research examining the impact of
lean system on firm performance to bring to light shortcomings in the body of literature.
In section five, the literature of lean system in services is reviewed with critical
assessment to help position the current study within this body of literature and to highlight
drawbacks weakening this literature. The penultimate section critically reviews the
literature focusing on the role of MAS and business strategy in the lean context to expose
its shortcomings. Finally, the last section in this chapter concludes by summarising the
main observations and explicitly articulating the research gaps.
2.2 The evolution of the lean system
Lean system was first implemented in Japan in 1950s in the automotive industry (Scott
and Walton, 2010, Lee et al., 2008), and more specifically at Toyota Production
Corporation (Liker and Morgan, 2006; Atkinson, 2004). The system was invented by
Taiichi Ohno who combined the advantages of both handcraft production and mass
production to establish the new developed management and production system, Toyota
Production System (TPS) or lean (Holweg, 2007; Lewis, 2000; Womack et al., 1990). The
system became of interest after the results of the International Motor Vehicle Program
15
were published in 1980s which highlighted the superior performance of Toyota in terms of
productivity and waste reduction at all levels compared to its counterparts of Western car
manufacturers (Holweg, 2007; Lewis, 2000). The term “Lean” was introduced by Jone
Krafcik (1988) and later, in 1990, was popularised by Womack et al. (1990) in their
seminal book “The machine that changed the world” (Worley and Doolen, 2006). Since
then, a great deal of research has been published to clarify the concept and investigate the
applicability of the Japanese system to other countries (Meier and Forrester, 2002;
Ahlstrom, 1998; Oliver et al., 1994).
However, the increasing interest in lean system has led to a strong debate in the literature
on what lean manufacturing is. For instance, some researchers believe that lean system is
just an extended model of the well known Japanese system, Just-in-Time (JIT) (Fullerton
and Wempe, 2009; Gorman et al., 2009; Smeds, 1994). On the other hand, Alagaraja
(2010) supports the notion that lean system is a developed version of Total Quality
Management (TQM). Similarly, Dahlgaard and Dahlgaard-Park (2006) review the relevant
literature related to lean, TQM, and six sigma and emphasise that lean has the same origin
as TQM and its practices should be viewed as supportive to the aim of TQM rather than as
an alternative. Contrasting the above-mentioned findings, Pettersen (2009) reports that
lean system is significantly different from its closest relative TQM leading to the
conclusion that lean system is a management concept on its own. Mitigating the debate,
Comm and Mathaisel (2005) and Radnor and Boaden (2008) state that when broken into
individual parts, the lean system is not new, but as a holistic approach it can be considered
as a new system. This statement is in line with the findings of critical research conducted
by Shah and Ward (2003) who find lean system to consist of four bundles, namely JIT,
TQM, Human Resource Management (HRM) and Total Preventive Maintenance (TPM)
where each bundle has its own practices.
In an attempt to crystallise the concept of lean system, several definitions have been
reported, each of which represents the authors’ perspective and understanding of the
system. For instance, Hinterhuber (1994, P. 275) defines lean system as an “integrated set
of attitudes and decision and action methods with which a firm can achieve sustainable
16
competitive advantage and increase its value in a turbulent environment”. Cooper and
Mohapeersingh (2008, P. 110) state that “lean thinking is a process-based method that
considers the interactions across the whole supply chain”. While to Atkinson (2004, P. 18)
lean system is “a concept, a process and a set of tools, techniques and methodologies that
can be used to attain and maintain effective resource allocation”. However, more
insightful definitions of lean system are provided by Shah and Ward (2007, 2003) and
Allway and Corbett (2002). Allway and Corbett (2002, P.45) consider lean system as “an
approach focusing on eliminating non-value added activities from processes by applying a
robust set of performance change tools, and emphasising excellence in operations to
deliver superior customer services”. Shah and Ward (2003, P.129) perceive lean system as
“a multi-dimensional approach that encompasses a wide variety of management practices,
including JIT, quality systems, work teams, cellular manufacturing, supplier management,
etc. in an integrated system”. Four years after their first definition, Shah and Ward (2007,
P.791) proposed that lean system is “an integrated socio-technical system whose main
objective is to eliminate waste by concurrently reducing or minimising supplier, customer,
and internal variability”. Given the variety of definitions provided, it seems unsurprising
to hear claims that there is still no agreement on a precise definition of lean system that
leads to some confusion when lean is considered for application (Kosuge et al., 2010; Jain
and Lyons, 2009). Despite such claims, viewing lean system as a multi-dimensional,
socio-technical system with several practices has been commonly acknowledged in the
recent literature of lean manufacturing (Dabhilkar and Åhlström, 2013; Furlan et al., 2011;
Birdi et al., 2008; Das and Jayaram, 2007). This perspective has also been adopted in the
current research.
2.3 The structure and mechanism of lean system
To facilitate the understanding and implementation process of lean system and guide all
companies across sectors in their journey towards being lean (Piercy and Rich, 2009),
Womack and Jones (1996) provide practitioners and researchers with five general
principles of lean that are based on the underlying assumption that all organisations are
consisted of processes (Burgess et al., 2009):
i. Value: to identify what customers value and run the business accordingly;
17
ii. Value stream: to identify all activities required to produce a product or service
whether adding or non-adding value;
iii. Flow: to ensure that adding-value activities necessary to produce and deliver a
product or service flow without interruptions;
iv. Pull: to produce according to customers demand; and;
v. Perfection: to continuously seek improvements to the process.
In addition to the five principles introduced by Womack and Jones (1996), several
researchers have provided their own roadmap of the transformation process from
traditional systems to lean system. In that direction, Allway and Corbett (2002) describe
the transformation process to be in five phases as follows:
Phase1: Assessment of the current state to identify weaknesses and strengths of the
current processes.
Phase 2: Determining the target state to be as a guideline for needed improvements.
Phase 3: Stabilising the operations by focusing on root causes of weaknesses and
inefficiencies in the processes.
Phase 4: Optimising the opportunities by introducing and applying methods and tools
to eliminate the root causes of problems.
Phase 5: Institutionalising the lean approach through continuous improvements of
processes to achieve the planned aims.
Atkinson (2004) proposes a more technical four-step model to consider for the
transformation process to lean:
Step 1: Selling and communicating the lean philosophy where the focus should be on
the results rather than focusing on the use of particular tools.
Step 2: Senior management commitment: where this commitment is crucial to
successful implementation of the lean concept.
Step 3: Design of projects: requires considering some aspects necessary to the lean
concept (e.g. Cross functional team, employees involvement).
18
Step 4: Selling the benefits of lean thinking.
In spite of the above presented models, the five principles introduced by Womack and
Jones (1996) have been relied upon by many researchers and organisations as they are the
keys to achieve benefits in operational and financial performance including customer
satisfaction (Vlachos and Bogdanovic, 2013; Fillingham, 2007; Endsley et al., 2006;
Lazarus and Andell, 2006; Jones et al. 1999).
These five famous principles indicate that the starting point of lean system is to focus on
and deeply understand the value from a customer perspective (Vlachos and Bogdanovic,
2013; Malladi et al., 2010; Comm and Mathaisel, 2000). This value can be created,
according to Hines et al. (2004), in two ways: (1) by eradicating wasteful activities from
process leading to a reduction in the associated cost and resources, and (2) by providing
better products and/or additional services at the same price leading to better value
perceived by customers. Adopting the first method to create value for customers, lean
system perceives any activity that does not add value to a product or service from a
customer perspective as completely waste and should be eliminated or controlled to
minimum (Petersen and Wohlin, 2010; Turesky and Connell, 2010; Grasso, 2005). For
some, the term waste, defined as any activity that does not add value, might seem vague.
Therefore, some researchers have made some attempts to classify the general term “waste”
into smaller categories so that it would be better understood and easier to trace. Ohno
(1988) provides seven categories of waste and Bodek (2007), in a more recent study, adds
one more to end up with a list of eight types of waste that are presumed to be applicable to
any area including office and administrative areas (Garrett and Lee, 2010). These eight
types of waste include:
a) Overproduction: to produce more than needed.
b) Unnecessary inventory: to build up and maintain any kind of inventory to a higher
level than needed.
c) Unnecessary motion: movement of staff that does not add value to a product from
customer perspective.
d) Transporting: unnecessary movement of materials.
19
e) Defects: products that do not satisfy quality standards; scrap and rework in
production.
f) Inappropriate processing: unnecessary processes that do not add value or not
required by a customer.
g) Waiting: idle time spent by employees waiting to be able to process again.
h) Unused creativity: the lack of efficient use of humanity skills.
The identification of waste in a process is not the end point in the lean journey. Rather, it
should be followed by actions to contain the avoidable waste (Ehrlich, 2006).
Consequently, some researchers view lean system to consist of two levels: lean philosophy
and lean practices (Bhasin and Burcher, 2006). While to others it is a three-level system,
lean philosophy, lean principles and lean practices (Arlbjørn et al., 2011). The philosophy
of lean focuses on the value from customer perspective and eliminating what exist from
the eight categories of waste presented above to enhance that value. The second level
involves the five principles introduced by Womack and Jones (1996) which can be a
roadmap to achieve the first level. While the third level, practices, tools and techniques, is
necessary to reflect the previous two levels practically and as a result, it represents the
actions to be taken (Bhasin and Burcher, 2006). Others, for example Hines at al. (2004)
perceive lean system as a two-level system involving strategic and operational levels. The
strategic level concerns the focus on what customers are willing to pay for, while the
operational level is related to the application of lean tools and techniques to approach the
aim of the strategic level.
Using the lean philosophy to focus attention on customer value, lean principles to guide
the transformation process and lean practices to identify and eliminate waste, lean system
is argued to generate several benefits to all adopters whether in manufacturing or services
(Shah and Ward, 2007, 2003). This claim has been subjected to considerable empirical
examinations for validation purposes as presented in the next section.
2.4 Empirical literature on the lean-performance association
As mentioned before, lean manufacturing has been studied extensively in the academic
literature since its invention. Various types of research have been used to report on this
20
system including conceptual studies (e.g. Cooper and Maskell, 2008; Dankbaar, 1997;
Womack and Jones, 1996; Cusumano, 1994), case studies (e.g. Lasa et al., 2009; Kennedy
and Widener, 2008; Karlsson and Ahlstrom, 1996, Oliver et al., 1994), simulation studies
(e.g. Li et al., 2012; Meade et al., 2010), and survey studies (e.g. De Menezes et al., 2010;
Fullerton and Wempe, 2009; Shah and Ward, 2007, 2003; Cua et al., 2001). However,
given the aim of lean manufacturing to improve adopters’ operations by increasing the
value delivered to customers through eliminating non-value added activities (Womack and
Jones, 1996), it is unsurprising that a considerable deal of attention has been devoted to
examine its effectiveness.
Before moving on to present the literature focusing on the lean-performance association,
there is an important aspect that deserves some attention. As indicated by the definitions
provided by Shah and Ward (2003) and Allway and Corbett (2002), lean manufacturing
involves a wide range of change practices that are usually implemented to achieve the aim
of lean manufacturing. Therefore, although lean manufacturing has been expressed as a
two or three-level system, the change practices have always been considered as the
practical weapons to attain the aim of the other levels. Consequently, the vast majority of
researchers have investigated the possible effect of lean practices (rather than its
philosophy or principles) on firm performance assuming these practices reflect the other
levels of lean manufacturing (Shah and Ward, 2003).
2.4.1 The effect of isolated lean practices on firm performance
When studying the impact of lean manufacturing practices on performance, some
researchers have focused on the isolated effect of individual practices (Pont et al., 2008).
For instance, to evaluate the effect of lean practices as well as the effect of using non
financial measures on companies’ profitability, Fullerton and Wempe (2009) conducted
their empirical research with a sample size of 121 manufacturing executives. Three lean
practices were considered, namely setup reduction, cellular manufacturing and quality
improvement. The results indicated that lean practices had a positive relationship with the
use of non financial measures. The important finding of this research was proving that the
utilisation of non financial measures mediates the relationship between lean practices and
21
firms’ profitability. Samson and Terziovski (1999) assessed the effect of six lean practices
(i.e. leadership, people management, customer focus, strategic planning, information
analysis, and process management) on operational performance of 1024 manufacturing
companies. The findings of this study showed a positive impact of HRM practices on
operational performance measured in terms of customer satisfaction, employee morale,
productivity, quality of output and delivery. However, no impact could be captured for
strategic planning and process management while information analysis negatively affected
operational performance of adopters. Similarly, Kaynak (2003) investigated the
interdependence between seven lean practices (i.e. management leadership, training,
employee relations, supplier quality management, quality data and reporting, process
management, and product/service design) and their impact on operational and financial
performance of 214 firms of which 85% are manufacturing firms while only 15% are
service firms. The author found that the use of HRM practices led to increase in the use of
TQM practices which then improved operational and financial performance.
Further, Powell (1995) used data from 54 service and manufacturing firms to examine the
impact of 17 HRM and TQM practices on financial performance measured in terms of
profitability, sales growth and an overall index. Although the findings indicated a superior
performance of TQM adopters over non-TQM adopters, only three individual practices
were significantly related to performance, namely committed leadership, open
organisation and employee empowerment. This credits the social practices over the
technical ones for improving financial performance. Talib et al. (2013) reported on the
impact of 17 lean practices on quality performance based on data collected from 172
service firms. Out of the 17 practices, training and education, quality system,
benchmarking, quality culture and teamwork were proved to have relation with quality
performance indicating superiority of the soft practices in affecting performance of service
firms. Bonavia and Marin (2006) provided evidence on the influence of 11 lean practices
on the operational performance of 76 manufacturing companies. Overall, the results
revealed no relationship between the extent of use of lean practices and improvement in
operational performance measured as internal quality, productivity, total stock and lead
time. In a similar vein, using a sample of 143 UK manufacturing firms and 135 UK
22
service firms, Alsamdi et al. (2012) found that the 10 lean practices used in the study were
employed similarly by service and manufacturing firms except for three namely, supplier
feedback, set up time reduction and total productive maintenance. These practices were
implemented more in the manufacturing context while employee and customer
involvement were found to be implemented more in the service firms. Moreover, lean
practices individually and collectively were found to have a significant association with
the performance of manufacturing firms. For service firms, lean practices as a whole was
significantly correlated with performance while individually only three (i.e. supplier
development, set up time reduction and total productive maintenance) out of ten practices
did not have a significant relationship with performance.
2.4.2 The effect of lean bundles
Kim et al. (2012) and Shah and Ward (2003) criticised the literature in the previous
subsection (2.4.1) on the grounds that lean practices are interdependent, and therefore
focusing on individual practices can be misleading. As such, they proposed that lean
practices should be classified into and examined as sets of internally consistent groups of
practices. Thus, Shah and Ward (2003) formally defined lean system as consisting of four
bundles, namely JIT, TQM, HRM and TPM, where each bundle has its own practices. As
a result, a new stream of research emerged which concentrated on the performance impact
of lean bundles rather than individual practices comprising each bundle (e.g. Agarwal et
al., 2013; Bonavia and Marin-Garcia, 2011; Rahman et al., 2010; De Menezes and Wood,
2006; Shah and Ward, 2003).
Among researchers in this stream of literature, Sakakibara et al. (1997) surveyed 41
plants in the manufacturing sector (i.e. transportation components, electronics and
machinery industries) to study the relationship between JIT practices and its infrastructure
practices with operational performance. The analysis indicated that JIT practices did not
directly relate to operational performance. Moreover, the results showed that the previous
results were changeable when introducing the infrastructure practices to the analysis.
More important was the result indicating that the infrastructure practices without JIT
practices had a direct effect on the operational performance. Cua et al. (2001) examined
23
the relationship of joint implementation of practices of three lean bundles (i.e. TQM, JIT,
and TPM) and manufacturing performance. Based on a sample of 163 manufacturing
plants in four countries, the authors found that manufacturing plants characterised with
higher manufacturing performance were associated with higher level of joint
implementation of both common and unique practices of TQM, JIT, and TPM. In addition,
plants applying a combination of manufacturing practices of the three programs had
higher manufacturing performance than plants focusing on only one program. However,
although the results supported a positive relationship between lean practices and
manufacturing performance, it was important to realise that different practices had
different effect on different performance dimensions (i.e. quality, on-time delivery,
flexibility and cost efficiency). Along this line, Shah and Ward (2003) conducted a survey
study of 1757 manufacturing plants to study the effect of lean system represented by four
bundles, namely JIT, TQM, TPM and HRM, on the operational performance represented
by 5-year changes in manufacturing cycle time, scrap and rework cost, labour
productivity, unit manufacturing cost, first pass yield and customer lead time. Each of the
lean bundles consisted of several practices. The results indicated a positive effect of each
of the lean bundles on operational performance.
Further, Pont et al. (2008) reported on the effect of three lean bundles (i.e. JIT, TQM and
HRM) on operational performance. Through surveying 266 plants located in nine
countries, the authors found a direct positive effect on operational performance (i.e.
quality, flexibility, on-time delivery and unit cost of manufacturing) by the JIT and TQM
bundles. However, the HRM bundle had its effect via the other two bundles. Rahman et al.
(2010) attempted to verify the purported positive impact of 13 lean technical practices
clustered into 3 factors on operational performance of 187 manufacturing firms. The
results of regression analysis proved the expected impact of the three lean factors on
operational performance. Based on data from 76 manufacturing firms, Bonavia and Marin-
Garcia (2011) investigated the effect of 4 HRM practices and a composite measure of 7
lean technical practices on 9 operational performance indicators. The results demonstrated
limited effectiveness of those practices given that 2 out of the nine discriminant models
were significant, namely when the stock level and productivity were the dependent
24
variables. In addition, Agarwal et al. (2013) used data from 152 manufacturing firms to
test the assumed positive impact of an index of lean practices on both operational and
financial indicators. The results proved the influence of lean index on merely some of the
performance indicators including sales, profit and profit margin.
2.4.3 The synergy between lean bundles
Shah and Ward (2003) proposed that the four lean bundles (i.e. JIT, TQM, TPM and
HRM) could not only have an additive impact on firm performance, but also non-additive
impact stemming from the expected synergy (also called interaction) among these
bundles. However, the authors did not examine explicitly this notion. Only a few
researchers to date have devoted their efforts to provide empirical evidence on the
presence of synergy between the different lean bundles. These are discussed below.
Using data from 433 manufacturing and service firms, Cappelli and Neumark (2001)
examined the impact of TQM, HRM and their synergy on operational performance. The
results neither supported the effect of TQM on performance nor the synergistic effect of
TQM and HRM on performance. The authors demonstrated that HRM practices did not
improve productivity as no statistical relation could be detected between productivity and
all but two practices, namely computer use by workers and job rotation. Computer use by
workers was proved to be positively associated with productivity while job rotation was
found to have an inverse relation with productivity. In addition, some of the HRM
practices were evident to increase labour costs as they had a statistical positive relation
with labour costs. Patterson et al. (2004) investigated the impact of Advanced
Manufacturing Technology (AMT), TQM, JIT and HRM on firm productivity and profits
based on data from 80 manufacturing firms. The authors demonstrated a positive impact
of HRM on both productivity and profits. However, among the three technical bundles,
only AMT was revealed to be positively related to productivity but not to profits.
Moreover, no synergistic effect was detected in this study. Among the 10 interactions
tested, only the one between AMT and job enrichment was significant, but
disappointingly, indicating that AMT has stronger effect on productivity when job
enrichment is low. Birdi et al. (2008) focused on the relationships between HRM (i.e.
25
empowerment, training and teamwork), TQM, JIT and AMT and the productivity of
manufacturing firms. Their results proved a positive association between HRM and
productivity. However, the technical side of lean system (i.e. TQM, JIT, and AMT) did
not have any significant influence on productivity. Moreover, these authors detected
positive interactions among only some of the variables tested. Specifically, they found a
positive synergy between teamwork and training, teamwork and empowerment, TQM and
JIT. No synergy between training and empowerment, teamwork and training and
empowerment, AMT and JIT, TQM and AMT, AMT and TQM and JIT was captured.
Further, Furlan et al. (2011) examined the synergy between JIT and TQM bundles and the
effect of HRM bundle on that synergy in improving operational performance of 266
manufacturing firms. The authors found that the expected synergy between TQM and JIT
could be captured when HRM was high while under low values of HRM the synergy
diminished. Dabhilkar and Åhlström (2013) investigated the synergy between a set of
technical lean practices and a set of HRM practices by employing data from 127
manufacturing firms. The authors demonstrated a lack of synergy and lack of positive
impact of HRM practices on operational performance indicators. However, the results
supported a full mediation of the HRM impact on operational performance by the set of
lean technical practices. Das and Jayaram (2007) adopted the socio-technical perspective
to examine the synergy between four lean technical practices (i.e. kanban, group
technology, JIT supply, TPM) and three HRM practices (i.e. cross-trained employees,
operator teams, decentralised decision-making). Based on data from 322 manufacturing
firms, the authors proved the expected synergy between the two sets of practices on
operational performance. Flynn et al. (1995) studied the relationship between JIT and
TQM practices and their synergistic effect on performance. Using data collected on 42 US
plants, the results indicated that the infrastructure practices alone (information feedback,
plant environment, management support, supplier relationship, and workforce
management) contributed sufficiently to JIT performance represented by average cycle
time. Unique JIT practices also improved cycle time significantly but the addition of
unique TQM practices did not provide significant improvement to the cycle time. On the
other hand, quality performance was significantly explained by the infrastructure practices
26
but not by neither unique JIT practices nor unique TQM practices. Although unique TQM
practices did not contribute significantly and individually to JIT performance, they did not
improve cycle time through their interaction with infrastructure practices and unique JIT
practices and unique TQM practices. Similar results were reported on the effect of all of
unique JIT and TQM practices and infrastructure practices by their interaction on quality
performance. Using data from 1024 manufacturing firms, Challis et al. (2002) examined
the effect of only technical practices in three lean bundles (i.e. JIT, TQM, AMT) and their
interactions on employee and manufacturing performance. The results indicated a positive
influence of the three bundles on employee performance but only JIT and TQM were
proved significant in improving manufacturing performance. In addition, AMT was found
to moderate the relationship between JIT and TQM and both employee and manufacturing
performance. However, TQM was evident to enable AMT to positively impact both facets
of performance. Table 2.1 presents a summary of the literature on the lean-performance
association.
27
Table 2-1: Summary of the lean-performance association literature
Focus of the articles
Performance Additive Interaction
Article Sample
size Industry
Statistical
analysis lean practices/bundles Operational Financial
Individual
practices Bundles
Individual
practices Bundles
Fullerton and
Wempe (2009)
121 Manufacturing SEM Setup time reduction, cellular
manufacturing, quality improvement,
Shop-floor employee involvement -
ROS
*
Samson and
Terziovski
(1999)
1024 Manufacturing OLS
regression
Leadership, people management,
customer focus, strategic planning,
information analysis, and process
management
Customer satisfaction,
employee morale,
productivity, quality of
output, delivery
- *
Kaynak (2003) 214 85% manufacturing
firms and 15%
service firms
SEM Management leadership, training,
employee relations, supplier quality
management, quality data and
reporting, process management, and
product/service design
Market share, market share
growth, product/service
quality, productivity, cost
of scrap and rework,
delivery lead-time of
purchased materials,
delivery lead-time of
finished products/service,
purchased material
turnover, total inventory
turnover
ROI, sales
growth, profit
growth
*
Powell (1995) 54 Service and
manufacturing
Correlation
analysis
Executive commitment, adopting the
philosophy, closer to customers, closer
to suppliers, benchmarking, training,
open organization, employee
empowerment, zero-defect mentality,
flexible manufacturing, process
improvement, measurement.
-
Profitability,
sales growth
and an overall
performance
index
*
28
Talib et al.
(2013)
172 Service OLS
regression
Top-management commitment,
Customer focus, Training and
education, Continuous improvement
and innovation, Supplier management,
Employee involvement, Information
and analysis, Process management,
Quality systems, Benchmarking,
Quality culture, Human resource
management, Strategic planning,
Employee encouragement, Teamwork,
Communication, Product and service
design
Quality performance
(product, process and
service quality, employee
service quality, employee
satisfaction, customer
satisfaction, supplier
performance) - *
Bonavia and
Marin (2006)
76 Manufacturing Friedman’s
non-
parametric
test,
Wilcoxon
tests,
Mantel-
Haenszel
common
odds ratio
Group technology, pull system, quick
set up time, multi-functional
employees, visual factory-graphs or
panels, visual factory-housekeeping,
SPC, group suggestions program
(quality circle), standardisation, TPM,
quality controls
Internal quality,
productivity, total stock,
lead time
- *
Alsamdi et al.
(2012)
278 Service and
manufacturing
OLS
regression
Supplier feedback, JIT delivery by
suppliers, Supplier development,
Customer involvement, Pull,
Continuous flow, Set-up time
reduction, TPM, SPC, Employee
involvement
Customer satisfaction,
employee satisfaction,
cycle time, market share
and labour productivity
ROA,
production
cost *
Sakakibara et al.
(1997)
41 Manufacturing Canonical
correlation
analysis
Infrastructure bundle (Quality
management, work force management,
manufacturing strategy, organisational
characteristics, product design), JIT
bundle (setup time reduction, schedule
flexibility, maintenance, equipment
layout, kanban, JIT supplier relation)
Inventory turnover, on-time
delivery, lead time, cycle
time, flexibility, quality,
cost - *
29
Cua et al.
(2001)
163 Manufacturing Discriminan
t analysis
TQM bundle (cross-functional product
design, process management, supplier
quality management, customer
involvement), JIT bundle (setup time
reduction, pull system, JIT delivery by
supplier, equipment layout, daily
schedule adherence), TPM bundle
(autonomous and planned
maintenance, technology emphasis,
proprietary equipment development),
common practices bundle (committed
leadership, strategic planning, cross-
functional training, employee
involvement, information and
feedback).
Unit cost, quality, delivery,
flexibility, weighted
performance
- *
Shah and Ward
(2003)
1757 Manufacturing Hierarchical
regression
analysis
TQM bundle (New process equipment
or technologies, Competitive
benchmarking, Quality management
programs, Total quality management,
Process capability measurements,
Formal continuous improvement
program), JIT bundle (Lot size
reductions, JIT/continuous flow
production, Pull system, Cellular
manufacturing, Cycle time reductions,
Focused factory production systems,
Agile manufacturing strategies, Quick
changeover techniques,
Bottleneck/constraint removal,
Reengineered production processes),
TPM bundle (Predictive or preventive
maintenance, Maintenance
optimization, Safety improvement
programs, Planning and scheduling
strategies, New process equipment or
technologies), HRM bundle (Self-
directed work teams, Flexible, cross-
functional workforce)
Cycle time, scrap and
rework cost, labour
productivity, unit
manufacturing cost, first
pass yield and customer
lead time
- *
30
Pont et al.
(2008)
266 Manufacturing SEM JIT bundle (production planning,
facility layout, JIT material delivery,
JIT, kanban, setup time reduction,
small lot sizes), TQM bundle
(proprietary equipment, statistical
quality control, 5Ss, quality circles,
error proofing), HRM bundle (team
working, employee involvement, flat
organisation, training, continuous
improvement)
Quality, flexibility, on-time
delivery and unit cost of
manufacturing
- *
Rahman et al.
(2010)
187 Manufacturing OLS
regression
JIT bundle (Reduction of inventory,
Preventive maintenance, Cycle time
reduction, Use of new process
technology, Use of quick change-over
techniques, Reducing set-up time),
Waste minimization bundle (Eliminate
waste, Use of error proofing
techniques, Using pull-based
production system (Kanban),
Removing bottlenecks), Flow
management bundle (Reducing
production lot size, Focusing on single
supplier, Continuous/one piece flow)
Quick delivery, unit of
manufacturing cost,
productivity, customer
satisfaction
- *
Bonavia and
Marin-Garcia
(2011)
76 Manufacturing ANOVA,
discriminant
analysis
training, employment security, internal
promotion, contingent remuneration,
lean index (visual factory-graphs or
panels, visual factory-housekeeping,
SPC, group suggestions program
(quality circle), standardisation, TPM,
quality controls, quick setup time)
Internal quality,
productivity, total stock,
lead time, minimum batch
size, on-time delivery,
employee turnover,
employee absenteeism
- *
31
Agarwal et al.
(2013)
152 Manufacturing Panel data,
OLS
regression
Operations management bundle
(Adoption of Lean Manufacturing,
Rationale for the adoption, Process
problem documentation, Operations
Performance tracking, Operations
Performance review, Operations
Performance dialogue, Consequence
management), Performance
management bundle (Types of goals,
Interconnection of goals, Time
horizon, Setting stretch goals, Clarity
of goals), people management bundle
(Instilling a talent mindset, Rewarding
top performance, Addressing poor
performance, Promoting high
performers, Attracting high
performers, Retaining high
performers)
Labor productivity (sales
per employee, profit per
employee)
Sales, profit,
profit
margin,
ROE, 1 year
sales growth
*
Cappelli and
Neumark (2001)
433 Service and
manufacturing
Panel data TQM, self-managed teams, regularly
scheduled meeting, teamwork training,
job rotation, cross training, pay-for-
skill programs, gain sharing,
benchmarking, computer use
Sales per worker, total
labor costs per worker, the
ratio of those two measures - *
32
Patterson et al.
(2004)
80 Manufacturing Sequential
regression
AMT (computer-numerically
controlled machines; flexible
manufacturing systems; computer-
aided design and engineering;
manufacturing resource planning; and
any industry-specific equipment),
TQM (quality of suppliers and
incoming materials; methods of
quality assurance applied to the
company’s production processes and
own end-products; the extent of use of
various techniques such as statistical
process control and quality circles; the
extent of training in quality for various
staff groups and measures to evaluate
quality and how they were used,
including procedures to feed back
quality performance to the workforce),
JIT (reductions in product-lead times;
methods to reduce set-up times; means
of limiting the amount of work in
progress; JIT purchasing initiatives;
ways of limiting the volume of
finished goods in stock; preventative
maintenance procedures; and kanban),
empowerment (job enrichment, skills
enhancement)
Labor productivity (sales
per employee)
Profit (sales-
costs per
employee
before profit
tax)
*
Birdi et al.
(2008)
308 Manufacturing Multilevel
analysis
Empowerment: employee
empowerment. Extensive training.
Team-based working. Total quality
management (Such practices include
Kaizen and continuous improvement).
Just-in-time production. Advanced
manufacturing technology (such as
CAD, CAM, computer-integrated
manufacturing and flexible
manufacturing systems). Supply-chain
partnering.
Company productivity
- *
33
Furlan et al.
(2011)
266 Manufacturing ANOVA,
Tukey test,
OLS
regression, F
test
JIT(production scheduling, facility
layout, small lots, JIT delivery,
kanban, setup time reduction), TQM
(5s, SPC, small group sessions,
mistake proofing), HRM (team
working, employee involvement, flat
organisation, training, quality circle,
continuous improvement)
Quality, dependability,
speed, flexibility, cost
- *
Dabhilkar and
Åhlström
(2013)
127 Manufacturing Hierarchical
regression
Lean bundle (work in progress
reduction, Setup times reduction,
Machine downtime reduction,
transportation time reduction, quality
system, JIT, pull system), HRM (team
work, continuous improvements,
quality control).
Productivity,
manufacturing quality, on-
time delivery, lead time
- *
Das and
Jayaram (2007)
322 Manufacturing OLS
regression
kanban, group technology, JIT supply,
TPM, cross-trained employees,
operator teams, decentralized decision-
making
Cost reduction
performance, Quality
performance,
Manufacturing cycle time,
New product introduction
Time, Delivery
performance,
Customization
responsiveness.
- *
Flynn et al.
(1995)
42 Manufacturing Hierarchical
regression
Plant environment, Management
support, Supplier relationship,
Statistical process control, Product
design, Customer focus, Kanban, JIT
scheduling, Lot size reduction
Cycle time and perceived
quality
- *
Challis et al.
(2002)
1024 Manufacturing Correlation
analysis,
MANOVA,
OLS
regression
AMT, TQM, JIT Employee performance
(morale, productivity,
skills, etc.), manufacturing
performance (customer
satisfaction, cash flow,
total costs per unit, delivery
time, defects rate)
- *
34
2.4.4 Critical evaluation of the empirical literature on the lean-performance
association
The impact of lean system on firm performance has accounted for a large proportion of the
extant literature in the lean system field. However, a closer look at this body of literature
reveals some limitations that should be considered to further advance this literature. First,
as can be seen from table 2.1, in examining the impact of lean system on performance,
several researchers focused on the direct influence of individual practices in isolation, as
shown in the subsection 2.4.1. By doing so, these studies did not account for the likely
collaboration between those practices in affecting firm performance (Kim et al., 2012;
Shah and Ward, 2003). Consequently, these studies did not help to uncover the full
potential of lean system. In addition, researchers in this part of the literature used different
practices to represent lean system along with different measures of firm performance (De
Menezes et al., 2010; Shah and Ward, 2003). Therefore, given the limited focus of these
studies on individual practices of lean system in isolation and the use of different
measures, these studies reported inconclusive results making it difficult to conclude about
the effectiveness of lean system (De Menezes et al., 2010; Shah and Ward, 2003).
In an attempt to overcome shortcomings of the previous studies, a new stream of research
emerged that assessed the performance impact of internally consistent groups of lean
system practices (i.e. lean bundles) rather than that of individual practices included in each
bundle as recommended by Shah and Ward (2003). However, despite advancing our
knowledge on the performance effect of lean bundles, the new stream of research
(subsection 2.4.2) mainly examined the direct and additive effect of lean bundles while
ignoring the possible non-additive effect stemming from collaboration between lean
bundles to improve performance (Dabhilkar and Åhlström, 2013; Furlan et al., 2011).
Therefore, the full potential of lean system is still yet to be uncovered through more
rigorous studies. In addition, almost all studies in this part of literature devoted their
attention to the impact of lean bundles on operational performance. With the exception of
Agarwal’s et al. (2013) study, neither of the above studies has investigated the effect of
the same bundles on financial performance.
35
To date, only a few researchers have made advanced efforts to highlight and examine the
non-additive influence of lean bundles on firm performance as can be seen in subsection
2.4.3. Among these few attempts, some researchers focused merely on the synergy
between the technical bundles of lean system (e.g. Furlan et al., 2011; Challis et al., 2002).
However, other researchers who focused on the wider definition of lean as provided by
Shah and Ward (2003) used a limited number of practices to represent each bundle (e.g.
Dabhilkar and Åhlström, 2013; Birdi et al., 2008; Das and Jayaram, 2007; Patterson et al.,
2004). Given these characteristics of this emerging body of literature, research to date has
largely failed to provide a definitive statement on whether lean bundles interact together to
generate an impact on performance over and above that expected from each bundle
separately. Moreover, since the focus has shifted to lean bundles, the majority of studies
have been limited in their focus to the effect of lean bundles on some operational
indicators as are apparent in subsections 2.4.2 and 2.4.3. To be able to offer a stronger
statement on the effectiveness of lean bundles on performance, there is an urgent need for
empirical studies that examine not only the additive but also the non-additive effect of
lean bundles on operational and financial performance using a wider set of practices to
adequately represent lean bundles (Kim et al., 2012; Patterson et al., 2004; Kaynak, 2003;
Shah and Ward, 2003). Finally, it is apparent that the reviewed and presented research in
this section is biased towards investigating the lean-performance relationship in the
manufacturing context. With the studies of Talib et al. (2013) and Alsamdi et al. (2012) as
the exceptions, most researchers have focused on manufacturing including those who have
employed a mixed sample in their studies due to the noticeable bias in the structure of
their samples towards manufacturing firms (e.g. Kaynak, 2003; Powell, 1995). As a result,
the findings of the above studies do not help in examining the universality of lean
practices at sector level (i.e. manufacturing versus services).
2.5 Lean system in the service sector
It is clear from the preceding discussion that there is a large body of literature which has
investigated lean practices in the manufacturing operations. The question that arises at this
point is “what is about lean system in the service context?” More specifically, are lean
36
practices applicable to service operations? This section reviews the literature of lean
service with critical assessment to its credibility which assists in suggesting avenues to
improve the knowledge in this specific area.
2.5.1 The emergence of the lean service concept
The service sector has been growing rapidly in many developed countries (Chase and
Apte, 2007; Ellram et al., 2004). Manufacturing companies have started to focus on the
provision of services given the higher revenue generated from service operations and the
need to supplement manufacturing capabilities in order to improve their competitive
position (Wu and Wu, 2010; Chase and Apte, 2007; Voss, 2005). These facts together
necessitate that more consideration and control should be devoted to service operations
(Malmbrandt and Åhlström, 2013; Ellram et al., 2004). Consequently, attempts to apply
some concepts such as lean system developed in the manufacturing and which has been
assumed to deliver various benefits to adopters may result in significant outcome (Vlachos
and Bogdanovic, 2013; Apte and Goh, 2004).
The transference of lean practices to services is relatively recent compared to the long
history of applying lean practices in manufacturing. Womack and Jones (1996) formally
introduced the term lean thinking that expanded lean manufacturing to include non-
manufacturing processes indicating the applicability of lean system to processes other than
manufacturing. Despite that, the term lean service was introduced explicitly in the
academic literature in a pioneering article written by Bowen and Youndahl (1998) two
years after the term lean thinking was reported (Suárez-Barraza et al., 2012).
Two main reasons are behind the late introduction of lean to the service sector. First,
unlike manufacturing products, services are characterised by being intangible, perishable,
labour-intensive, heterogeneous (Soltani et al., 2012; Nie and Kellogg, 1999; Brignall et
al., 1991; Sasser, 1976), and most importantly the presence of customers during the
delivery of services, which is the major difference between lean manufacturing and lean
service (Abdi et al., 2006; Sampson and Froehle, 2006; Bowen and Youngdahl, 1998).
Those unique characteristics are argued to expose service operations managers to some
37
difficulties which do not face their counterparts in manufacturing (Soltani et al., 2012;
Sampson and Froehle, 2006; Nie and Kellogg, 1999). Mefford (1993) emphasises the
difficulties associated with properly defining quality in services in addition to the
variability inherent in services due to the labour intensity characterising most services.
Moreover, the convergence between the production and consumption of services resulting
from the presence of customers adds to that variability and to the difficulty of defining and
measuring quality (Soltani et al., 2012; Sampson and Froehle, 2006; Brignall et al., 1991).
As a result, manufacturing systems, tools and techniques are claimed to be insufficient and
inadequate to services (Sampson and Froehle, 2006; Nie and Kellogg, 1999). In a similar
vein, Piercy and Rich (2009) and Liker and Morgan (2006) underline the need for a higher
level of reconsideration when applying lean system into industries lacking the presence of
physical products.
Second, as shown in the introduction chapter, lean manufacturing has emerged as a result
of the changing market environment accompanied by an increasing level of competition
facing manufacturing firms (van Biema and Greenwald, 1997). Broadly speaking, service
providers have not experienced a similar level of competition to that faced by
manufacturing companies (van Biema and Greenwald, 1997). Thus, seeking more
effective production and management systems such as lean service has not been a priority;
thereby they lagged far behind their cousins in the manufacturing sector in considering
such a new system (Schmenner, 1986).
However, given the continuing changes in economic conditions, several researchers have
reported that the challenges of globalisation and accompanying competition have
approached the service sector (Karmarkar, 2004). Consequently, there has been a serious
need for service firms to reconsider their operations management methods (Vlachos and
Bogdanovic, 2013; Schmenner, 1986). In addition, in spite of the unique characteristics of
services reported above, lean system has been claimed to be relevant to services for at
least one reason. Lean system is argued to focus on processes rather than products and all
companies, manufacturing and non-manufacturing, are a compilation of processes that are
used to provide customers with products and/or services (Jimmerson et al., 2005; Allway
38
and Corbett, 2002). In line with this perspective, Abdi et al. (2006) have discussed the five
lean principles suggested by Womack and Jones (1996) and have explained the validity of
those principles from a service point of view. Armed with this argument, several
researchers have stressed the need and applicability of lean practices to services (e.g.
Kosuge et al., 2010; Endsley et al., 2006; Jones et al., 1999). Consequently, the literature
of lean service, whether in the public or private sector, is observed to be escalating. For
instance, Radnor (2010) finds that 51% of the publications on the public sector focus on
‘Lean’ and 35% of those are in the Health Services. Despite the focus on lean service in
healthcare, lean service can be found in other industries such as higher education,
retailing, banking and financial services, telecommunication, software, fast food, airlines
and others, as presented below.
2.5.2 Lean service in the healthcare sector
The literature of lean service has been dominated by applicability and implementation of
lean practices in healthcare (Vlachos and Bogdanovic, 2013; Holm and Ahlstrom, 2010).
Around 90 publications on lean service in healthcare were published in ten countries from
2002 onwards (Souza, 2009). 57% of these papers concerned the implementation of lean
in the US healthcare sector which might reflect the success of lean in private sector while
29% was related to lean in the UK healthcare sector indicating increasing popularity of the
concept in the public sector. That domination could be due in part to the importance of
healthcare since it deals with human’s life. Therefore, it is crucial to improve the
performance of that sector especially when realising that many activities in healthcare do
not directly add value from patients’ perspective (Fillingham, 2007). It could also be due
to the pressure exerted on the healthcare sector to seek methods to improve productivity,
efficiency, quality of services and cut costs in response to cutting the related budget by
associated governments (Radnor and Walley, 2008). In the following space, the literature
of lean service in the healthcare sector is summarised.
Manos et al. (2006) believed that the eight types of waste (presented in section 2.3)
originated in manufacturing did exist in healthcare although in slightly different forms.
Thus, lean service could be sought to eliminate or at least reduce the level of that waste
39
(Cooper and Mohabeersingh, 2008; King et al., 2006; Manos et al., 2006) so that
providing better care to patients. Ben-Tovim et al. (2007) stated that lean service might
have a major positive effect on healthcare in Australia and elsewhere. Manos et al. (2006)
supported the implementation of various lean techniques to the healthcare environment
and provided a list of 18 lean tools and techniques that, they believed, were as applicable
to healthcare as to manufacturing.
Cooper and Mohabeersingh (2008) reported on the necessity for the UK healthcare sector
to strive for minimising waste, improving efficiency, creating a harmonious working
environment. To attain this, they argued about the appropriateness of lean service.
Therefore, they studied the implementation of lean service in five functions of the
healthcare services and found it to improve efficiency; reduce waiting time and costs as
indicators of waste. Moreover, they appraised lean as a tool which could be strategically
significant in terms of cost reduction and achieving high turnaround using the same
resources but in a more effective way. In a similar vein, Burgess and Radnor (2010)
focused on the applicability and performance outcome of three lean practices, namely
process mapping, 5S and cross functional employees in two hospital trusts in the UK. The
results supported the applicability of lean practices and their capability in providing better
understanding of the process, improving stock control, improving the understanding and
resolution of the impact of variation, exposing problems and risk factors, reducing the “did
not attend” rate and improving employees satisfaction and morale.
Other successful applications of lean service practices were reported from Flinders
Medical Centre in Australia by King et al. (2006) and Ben-Tovim et al. (2007) in two
separate studies. The authors found that process mapping followed by redesigning
processes and the use of what is known in manufacturing as production cells led to
improvement in patients flow, reduction in operational costs, increase in the number of
patients treated, reduction in the overall time from a patient entry to discharge as well as in
the average number of patients in the Emergency Department at any time. Moreover, lean
implementation in Intermountain Health Centre was presented by Jimmerson et al. (2005).
The centre enjoyed a reduction of a significant amount of wasted time of front-line
40
workers, reduction in the amount of errors occurred and improvement in employees and
customers satisfaction with no or little investment.
More positive results on the powerful effect of lean service on hospitals processes were
attained by Bolton Hospitals NHS Trust (Fillingham, 2007). The program consisted of
determining the value from patient’s perspective, examining whether current processes
provide that value, redesigning the processes to eliminate non value-added activities,
facilitating the process flow and finally, making sure that the redesigned processes deliver
the wanted value by patients. To apply the program, some practices were needed such as
5S, value mapping, and visualisation. A 42% reduction in paperwork, an improvement in
team working, a 38% reduction in the process time, a reduction in the total length of stay
by 33% and a reduction of mortality by 36% were all achieved. Furthermore, Lodge and
Bamford (2008) employed the following lean practices in the radiology department of the
Pennine acute Hospitals NHS trust: automation, process simplification techniques and
training. As a result, waiting time was reduced by 30%, control was enhanced, better
understanding of capacity requirements was improved and access to patient information to
answer queries became easier.
Personal experience in the deployment of lean service was reported by Bushell et al.
(2002). The authors deployed four lean practices which were value stream mapping,
standardised operations, workplace organisation and visual/audio controls. The authors
were optimistic that lean implementation would improve the care delivery process at the
Progressive Healthcare. Poksinska (2010) reviewed the literature concerning the
implementation of lean production in health care. The results of the literature review
revealed that the benefits from applying the lean service could be reported at two levels.
The first level relates to the performance of the healthcare system including decreased
overall time spent by patient on care, increased number of patients handled, reduced
number of errors, reduced waiting times, increased patient and employees satisfaction,
reduction of overtime and inventory costs. The second level relates to the development of
employees and work environment including increased attention of employees to waste and
their attitude to problem solving and having more organised work environment.
41
Kim et al. (2006) forwarded one more successful story on the implementation of lean
service at Virginia Mason Medical Center (VMMC). This implementation resulted in a
cost reduction of approximately a half million dollars, increased profit margins and
improved space utilisation at its cancer centre resulting in a 57% increase in the number of
patients treated in the same allocated space. Added to that, Esain et al. (2008) reported on
the deployment of 5S as one of the lean service practices in a large NHS Trust. The
authors argued that 5S was a helpful tool for improving the understanding of current
processes and setting the foundations for changes although it had to be considered as a
starting point of the change process. Examining the annual report and other archival data
for 152 UK hospitals, Burgess et al. (2009) found that more than half (80 hospitals) of all
English hospitals articulated the use of lean service practices in their annual reports. Based
on this articulation, the authors classified the 80 hospitals into six categories. These
categories were adopting a systemic approach, multiple projects, a few projects, rapid
improvement events, productive ward only and ‘tentative’ to lean by piloting a small
project. Moreover, the authors found no evidence of the effect of lean implementation
approach on hospitals performance. However, this lack of evidence was attributed to
having a very small number of hospitals which considered a systematic approach in
adopting lean service.
2.5.3 Lean in other service sectors
Lean service is not confined to healthcare services among other service industries
although they have dominated the lean service literature (Vlachos and Bogdanovic, 2013;
Holm and Ahlstrom, 2010). A considerable number of researchers were found to be strong
believers in the validity of lean service to the various service processes. For instance, Abdi
et al. (2006) discussed the five lean principles suggested by Womack and Jones (1996)
and explained their validity from a service point of view. Similarly, Ehrlich (2006)
supported the applicability of lean principles such as the identification of customer value,
simplification of process and pull system to all types of processes. In addition, Jones et al.
(1999) claimed that waste reduction, employee empowerment and focusing on value-
adding activities from customer perspective were universally applicable. Moreover,
42
Comm (1999) argued that lean was successfully deployed by service providers such as
airlines, hotels and higher educations. Lee et al. (2008) referred in their article to the
successful implementation of lean concepts in some service companies such as Jefferson
Pilot Financial (a life insurance company), Zara (a Spanish clothier), and Fujitsu Services
(an international IT services company).
2.5.3.1 Lean service and IT industry
Staats et al. (2011) examined the validity of lean service in a software service firm, Wipro
Technologies. The authors found that lean service was capable of improving operational
performance such as processing time and labour productivity. This study was particularly
important because it proved that even in industries characterised by non-repetitive tasks
and a high level of uncertainty concerning customers demand; lean practices were still
applicable and yielded positive results. In a similar vein, Malladi et al. (2010) employed a
case study methodology to report on the implementation of lean service in the IT industry.
The results of their case study revealed that lean service improved efficiency which was
translated to savings leading to value creation to customers and the organisation.
2.5.3.2 Lean service and call centers
Piercy and Rich (2009) studied the applicability of lean service to pure service processes,
namely call centres of three financial service companies. The results of their study
conveyed a significant improvement in the performance of all call centres examined. Lean
service was found to reduce employee absenteeism, work in progress and process cycle
time. It freed staff time and related costs and improved employees’ morale and service
quality, thereby improving customer satisfaction. Sprigg and Jackson (2006) reported on
the effect of two lean service practices job-related strain among 836 call handlers from 36
call centres. The two practices were process simplification typified by dialog scripting and
work flow integration typified by performance monitoring. The results indicated a
negative effect of lean practices on job-related strain. Call handlers who were subject to
higher levels of dialog scripting and performance monitoring experience higher job-
related strain compared to others. Furthermore, the authors forwarded that this positive
relationship between job-related strain and lean practices was virtually fully accounted for
43
by work design characteristics such as lower control over work methods and timing, lower
task variety and skill utilisation, higher workload, role conflict and lower role clarity. In
another study focusing on lean service in call centres, Piercy and Rich (2009) reported on
the impact of three lean service practices (i.e. value stream mapping, process redesign and
continuous improvement) on the performance of three call centres. The findings showed
that lean practices led to a reduction in operational costs as well as an improvement in
customer satisfaction by virtue of the reduction in waiting time to complete a customer
request.
2.5.3.3 Lean service in the financial industry
The financial service industry was also found to be a good candidate for lean service given
the high percentage of waste inherent in its processes (Atkinson, 2004). Atkinson (2004),
through a study of eight functions in a 1200-person business in the financial sector, found
that 200 work activities represented source of waste in the form of reworking the same
activities often several times and that led to wastages in staff time quantified to about 40%
of labour cost. This particular result could be used to add support to the effective way of
making improvements by focusing on the deletion of non-value adding activities rather
than attempting to improve the value added activities (Fillingham, 2007). Based on this
finding, lean service practices could be used to eradicate such waste and improve
processes. Swank (2003) described the implementation process of lean service in a
financial service company where the results validated the findings of Atkinson (2004) and
highlighted the effectiveness of lean service in such environment. At the aim of reducing
process variability and improving its competitive advantage by providing customers with
better service quality, Jefferson Pilot Financial (a life insurance company) examined its
processes to identify major areas for improvements (Swank, 2003). Implementation of
process mapping technique revealed suffering from long processing time and unnecessary
costs. To overcome those problems the company initiated a lean service program which
included the following techniques: 1- processing applications in small lots 2-group
technology (placing linked processes near one another) 3- standardising procedures 4-
eliminating loop-backs ( where work returns to a previous step for further processing) 5-
Setting a Common Tempo (employing the takt-time concept) 6- balancing workloads 7-
44
segregating complexity (to cluster tasks of similar level of difficulty into separate groups
with their performance goals) 8- posting performance results. To measure the effect of the
program the company had to rely on new performance measures which suited lean concept
(Schonberger, 2008). The results indicated that the company reduced the average time
from receipt of premier partner application to issuance of a policy by 50%, reduced labour
cost by 26% and eliminated 40% of errors leading to reissuance. Based on that, service
quality improved; employee and customer satisfaction increased and the company’s
competitiveness enhanced.
Bhatia and Drew (2007) proclaimed the experience of a European bank with lean service
practices. The authors reported that the bank decreased time for mortgage applications
from 35 days to only 5 days. As a result, revenues and cost savings were increased.
Another successful story related to lean service in banks was forwarded by Yavas and
Yasin (2001). The authors argued that four tools developed in the manufacturing context
could be of high importance to banks. The use of root cause analysis, benchmarking,
process re-engineering and continuous improvement yielded remarkable benefits in the
bank they report on. The bank achieved an increase in customer satisfaction from 71% to
94%, a decrease in customer complaints by about 35%. In addition, waiting time for
customers automatically reduced and the bank goodwill in terms of introducing the bank
to other customers by the current customers increased from 68% to 89%.
2.5.3.4 Lean service and telecommunication industry
The telecommunication industry has also been invaded by lean service practices. Jones et
al. (1999) focused on the implementation of three lean tools in BT, a telecommunication
provider. The lean tools used were value stream analysis, root cause analysis and use of
new technologies. BT improved customer service, reduced operating costs, reduced
inventory and rework and increased the degree of flexibility to provide new services. In a
similar vein, a European telecommunication company benefited from the implementation
of some of the lean service practices (Bhatia and Drew, 2007). The company improved
productivity by 40% and reduced recurring failures by 50% (Bhatia and Drew, 2007).
45
2.5.3.5 Lean service and education
In the education industry, Hines and Lethbridge (2008) stated that lean thinking was
uncommon to university’s processes and activities although it could result in impressive
results. Comm and Mathaisel (2005 b) declared that institutions of higher education lacked
real understanding of lean service and struggled to identify their primary customer.
Therefore, higher education institutions were advised to educate employees on lean
concepts, follow the five lean principles as introduced by Womack and Jones (1996),
define appropriate metrics for success and concentrate on developing outsourcing,
collaboration programs, and technology initiatives (Comm and Mathaisel, 2005b).
Moreover, the same authors studied the suitability of lean system to higher education
institutions by descriptively examining 13 private and 5 public universities. The authors
believed that universities were good candidates for lean service due to being a compilation
of departments, offices and divisions. Therefore, lean practices could be relied upon to
eliminate waste form processes leading a university to focus on its main activity: teaching
and doing research. Maguad (2007) argued that several lean tools could be used to
eradicate different types of waste involved in the educational process. The author
suggested the following tools to be used: 5Ss, mistake-proofing, value stream mapping,
quick change over, self-inspection, TPM and kaizen. To facilitate the employment of those
tools, the author stressed the need for teamwork spirit and strong communication.
Following the procedure outlined above lean would improve workers understanding of
their crucial role in the organisation and waste would be eliminated. Tatikonda (2007)
recommended the implementation of lean techniques specifically quality function
deployment and cell layout to improve the quality of accounting education. He stated that
in 30 years of his experience, improvements of lean practices were not matched with any
other improvement systems. In a similar vein, Emiliani (2004) examined the applicability
of lean service to a graduate business course in order to improve the value delivered by the
course, eliminate waste and improve quality. The lean service program employed
consisted of the following practices: continuous improvement, 5Ss, JIT, load smoothing,
respect for people, standard work and visual control. The program yielded a significant
level of student satisfaction.
46
2.5.3.6 Lean service and housing services
Housing and construction services appear to be also appropriate for lean service practices.
McQuade (2008) contended that lean service practices could be effective in improving
business performance and responding to customers. In that context, the author
communicated the implementation of lean practices in Flagship Housing Group. Existing
processes in the company were examined and mapped to identify waste and value from
customers’ perspective. As a result of the new way of thinking, 80% waste was considered
from a customer point of view. Redesigning the processes taking into account what
customers value reduced failure demands from 70% to 30%, total time on repairs
decreased from 129 days to 7.7 days, one million pound operational savings were
achieved. Garrett and Lee (2010) examined the implementation of some of the lean tools
into the construction submittal process. VSM, 5s, poka yoke, JIT, visual controls and
failure mode and effect analysis were all implemented into different areas of the process.
The results of those tools were measurable and impressive. Lead time decreased by 40%,
process time improved by almost 25% and the number of activities decreased from 8 to 5.
2.5.3.7 Lean service and the distribution industry
Searcy (2009) described the lean initiatives conducted to expedite the performance of the
billing department of AMG, a distributor company. The company introduced employees to
the concept of lean through sessions of training, and then those employees were required
to conduct process mapping activity to visualise the process. The use of modern cost
accounting methods such as time-drive ABC helped the team to identify the amount of
waste inherent in the process. The root causes of waste were traced and quality circles
were established to find out how to eliminate them. The methodology followed by AMG
reduced the number of errors occurring in the billing department as well as freeing staff
time to be used in other departments. In addition, the results proved that training was
crucial in sending signals about the high management commitment to lean concept.
Communicating the initial results of lean transformation assisted in motivating employees
to maintain lean principles and believed in their effects. Rajiv (2009) stated that
47
implementing lean service to warehousing operation in his company yielded 8 million
savings over three years.
Value stream mapping and root cause analysis among other lean service practices were
highly powerful in identifying and understanding waste in the transportation process of
Mexican food distribution company (Villarreal et al., 2009). However, in contrast to the
norm of striving to apply small lots technique, the company combined a set of orders
together to improve vehicle capacity utilisation. The benefit of that action was estimated
to reduce shipments by 25% which might translate to 1.8 million peso per year. Safari
Park in the UK also enjoyed some benefits from employing lean service as reported by
Julien and Tjahjono (2009). The park sought to increase its profit by eliminating waste and
improving the efficiency of major activities while improving customer satisfaction.
Therefore, value stream mapping, automation and 5S were employed in the feeding
logistics at Safari Park. The deployment of those tools led to a more simplified process
involving less waste and considerable financial savings of about 91000 pounds. It should
be noted that involving employees was essential in the implementation process. Through
surveying employees in two logistic service providers, the author found that lean system
had a considerable positive effect on employees’ performance through offering a higher
level of autonomy and creativity and that will in turn improve the company effectiveness.
2.5.3.8 Lean service and office operations of manufacturing firms
More evidence on the applicability of lean practices to service operations could be
highlighted from the adoption of these practices by manufacturing firms to improve their
back offices operations. Lean system in general is designed to improve all activities in a
process rather than focusing on isolated activities (Petersen and Wohlin, 2010; Scott and
Walton, 2010; Arbos, 2002). Improving isolated activities simply means that problems or
waste may shift to the adjacent activities, thereby offsetting improvements already
achieved (Joosten et al., 2009; Hines and Lethbridge, 2008). Realising this fact, several
manufacturing organisations have attempted to extend their implementation of lean system
to non-manufacturing areas. For instance, Kato Engineering, a manufacturing company,
realised that it could benefit from lean system in its offices as it did in its production
48
operations (Tonya, 2004). The problem in this company was that the reduction in the
production cycle time achieved after the use of lean practices was offset by inefficiencies
in sales, change-orders and on-paper process management. The company relied on process
mapping and changing the physical layout as two lean practices. However, this attempt
required cross functional team, training, employees involvement. The results of this
implementation had been a reduction in the sales-order cycle time by 59%, in the
engineering change-order cycle time by 91%, response time to customer request by 83%
and employees errors by 69%. Rearranging work locations alone led to a reduction in the
total distance employees travelled per day from 1886 feet to 262 feet (Tonya, 2004).
Similarly, Wayne (2005) reported on the implementation of lean practices to improve the
performance of non-manufacturing areas at Bent River Machine after their successful
implementation in the manufacturing operations. The company employed the same set of
lean techniques used before in the manufacturing such as value stream mapping, 5S, visual
control and automation. The company realised the importance of training as well as
linking the benefits that the company would achieve to the benefits that employees would
achieve from successful lean effort. Consequently, the company enjoyed a reduction in
lead and processing time from 4 days to half a day, saved space and improved the
organisation of offices. Kennedy and Widener (2008) detailed the implementation process
of lean practices in the shop floor and accounting function of a manufacturing company,
EBS, a subsidiary of Tri-Metal. After the implementation of lean practices in its shop
floor, the company found it necessary that similar practices be applied in its accounting
function to sustain the results achieved. To that end, the company streamlined transactions
processes, ceased the use of standard costing in favour of actual costs, used kanban system
to facilitate purchasing and finally, and changed its performance measurement system.
That initiative resulted in freeing staff time in accounting function.
Reinertsen and Shaeffer (2005) supported the applicability of lean system to the R&D
function. However, for lean to be successful, the authors emphasised the importance of
understanding the differing nature between manufacturing and R&D functions. For
instance, eliminating variability in manufacturing processes was favourable but it was not
49
in R&D processes because part of that variability added value in reality. However,
employing the concept of small lots could easily be achieved and results in a reduction in
idle time. The use of pull instead of push concept combined with the deployment of small
lots could reduce queues and compress cycle time. In simple words, lean principles if
applied carefully would improve cost, quality and speed of the R&D process (Reinertsen
and Shaeffer, 2005). Haque and James-Moore (2004) addressed the suitability of lean
principles and practices to the new product development process in aerospace industry in
the UK. The findings of two case studies indicated that the five well-known principles of
lean system were compatible with the product development process. In addition, several
lean practices such as value stream mapping, takt time, standardisation, visual control, 5s,
single piece flow and 5 whys were found to be easily transferable to the process provided
that multi-functional team was present to facilitate the implementation of those practices.
2.5.4 Evaluation of the lean service literature
In subsections 2.5.2 and 2.5.3, studies which focused on the adoption and effectiveness of
lean service were reviewed. However, my intention was not to present all publications in
this area and I did not. Rather, I presented a sample of the 221 articles identified through a
systematic search for publications on lean service explained in detail in subsection 3.2.1 of
chapter 3. This sample of articles indicated a rising level of interest in lean service among
academics and practitioners and wide diffusion of its practices across the various service
industries. Classifying the whole population of lean service articles (221) per industry and
over time reinforced this conclusion. Figure 2-1 presents the classification of 221 articles
based on industry type. As shown in the Figure, healthcare and office operations have
been the most popular application areas for lean practices in the service sector. However,
Figure 2-1 also shows that lean service practices have not been totally uncommon to other
service industries. The trend of publications on lean service since 1993 is illustrated in
Figure 2-2. The Figure reveals an increasing interest in lean service among the academic
community with around 30 publications in each of 2011 and 2012.
However, despite the rising interest in lean service and its wide spread among various
service industries, this body of literature seems to suffer from two shortcomings. These
50
two shortcomings may hinder more advance knowledge to be acquired on lean service if
not taken into consideration.
From Figure 2-1, the first observed limitation of this literature is its strong focus on
healthcare and office operations at the expense of other areas especially hotels,
consultancy services, telecommunication, banking and financial service. However, this
observation is not limited to the lean service literature but it can be extended to the whole
service operations management research as documented by Machuca et al. (2007) in their
review of the associated literature. Therefore, more research in these areas is urgently
needed to have more sufficient information on the applicability and effectiveness of lean
service in these industries.
Figure 2-1: The classification of lean service literature per industry type
Healthcare
Office operations
General
Public sector
Education
Distribution, retail
and logistics
industries
Software
and IT
industries
Banking, financial
And insurance
Services
Call centers
Telecommunication
Consultancy
services Housing services
Hotel industry
51
Figure 2-2: The trend of lean service publications
The second limitation of the existing literature of lean service concerns the type of studies
dominating this literature. Synthesising information reported in section 2.5.2 and 2.5.3
reveals a significant reliance on conceptual (e.g. Allway and Corbett, 2002; Bowen and
Youngdahl, 1998) and case studies (e.g. Staats et al., 2011; Staats and Upton, 2011;
Swank, 2003; Arbos, 2002) to report on lean service. This observation supports previous
findings reported in two recent literature review studies conducted by Holm and Ahlstrom
(2010b) and Suárez-Barraza et al. (2012). Holm and Ahlstrom (2010b) have reviewed 56
articles on lean service and concluded that publications on lean service are dominated by
conceptual and case studies while most of them have focused on the healthcare processes.
Similar results have been reported by Suárez-Barraza et al. (2012) from reviewing 172
articles on lean service.
Conceptual and case studies are important for developing theory and generating research
questions about a specific phenomenon (Challis et al., 2002; Voss et al., 2002). However,
they suffer from inherent shortcomings that limit generalisation of their findings (Challis
et al., 2002; Voss et al., 2002). In addition, although case studies convey evidence from
real observations, they tend to be biased towards reporting positive findings which
necessitates caution when using their conclusions (Kepes et al., 2012; Challis et al., 2002).
This bias may justify the fact that the majority of case studies on lean service have
supported its effectiveness and little, if any, failure attempts have been published in the
0
5
10
15
20
25
30
35
199
3
199
4
199
6
199
7
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
201
2
Pu
bli
cati
on
nu
mb
er
Year
52
academic literature. Consequently, several researchers called for employing the survey
methodology to report on lean service in order to provide more generalisable findings
which enhances the knowledge on lean service and improves the quality of its literature
(e.g. Arlbjørn and Freytag, 2013; Malmbrandt and Åhlström, 2013; Suárez-Barraza et al.,
2012; Holm and Ahlstrom, 2010b).
The need for more survey studies on lean service is further reinforced by the findings
forwarded by Alsamdi et al. (2012) and Yasin et al. (2003). Alsamdi et al. (2012)
concluded that three lean practices (i.e. supplier feedback, set up time reduction and total
preventive maintenance) are more relevant to manufacturing than to services while it is the
reverse for employee and customer involvement. Yasin et al. (2003) fiound that
operations-oriented modifications including standardisation of operations, modifying
facility layout, increasing the level of automation neither decreased supplier-related
problems nor did they impact performance in service firms, but they did for manufacturing
firms. Thus, practices that are proved effective in manufacturing may not be in services.
Based on that, examining lean practices in the service context, through rigorous empirical
studies seems critical to advance the knowledge in this emerging area (Arlbjørn and
Freytag, 2013; Malmbrandt and Åhlström, 2013; Suárez-Barraza et al., 2012; Holm and
Ahlstrom, 2010b).
2.6 The literature on lean system, accounting system and business
strategy
As can be seen from the previous sections of this chapter, there has been a considerable
body of literature discussing and examining various aspects of lean system. This has
included a focus on defining lean system, highlighting its philosophy and principles,
identifying its practices and their interrelations, examining empirically the effectiveness of
lean practices on firm performance, and more recently exploring the applicability of lean
practices to other contexts than manufacturing. The question that can arise at this point
concerns the relationship between lean practices and the context in which they operate.
53
Lean practices have been suggested in response to some external contextual factors (i.e.
competition and changing market environment). However, considering merely the external
contextual factors can be misleading to adopters who may experience a disappointing
outcome accordingly. This happens when adopters or potential adopters do not have
proper understanding of the effect of internal contextual factors on the adoption level of
lean practices (Fullerton et al., 2013; Jain and Lyons, 2009; Shah and Ward, 2003). This
can also be supported by the contingency theory stating that any organisational,
management and operational system cannot be equally appropriate in all contexts and
environments (Shah and Ward, 2003; Drazin and Van de Ven, 1985). As a result, finding
out the conducive environment for lean practices can be highly significant for both theory
and practice given that such practices may not be equally implemented in all
organisational contexts (Fullerton et al., 2013; Malmbrandt and Åhlström, 2013; Shah and
Ward, 2003).
To date, the available literature providing information on the possible internal contextual
variables affecting lean practices can at best be described as sparse (Fullerton et al., 2013;
Malmbrandt and Åhlström, 2013; Voss, 2005; Shah and Ward, 2003). Within this stream
of literature, some researchers have highlighted the importance of management accounting
system (MAS) (e.g. Banker et al., 2008; Kennedy and Widener, 2008), business strategy
(e.g. Amoako-Gyampah and Acquaah, 2008; Ward et al., 2007), firm age and firm size
(Shah and Ward, 2003) as potentially influential factors on the implementation of lean
system.
2.6.1 Accounting system and lean system
The MAS is an integral element for most companies (Guilding et al., 2005; Mia and
Clarke, 1999). It is needed to provide accurate information on resources consumed by a
firm’s operations which can be used for different purposes including stock valuation,
pricing, make/buy decisions, operating performance evaluation, planning and supporting
improvement initiatives (Maiga and Jacobs, 2008; Fullerton and McWatters, 2004). The
role of the MAS in supporting other improvement initiatives has received a considerable
level of attention. This has been reflected by the amount of discussion and research
54
comparing the different types of MAS to understand their impact in the lean context (e.g.
Ruiz-de-Arbulo-Lopez et al., 2013; Schoute, 2011; Kennedy and Widener, 2008; Maskell
and Kennedy, 2007; Grasso, 2005; Ittner, 1999).
For decades, the traditional accounting system (TAS) (i.e. variable costing system and
absorption costing system) have played an important role in equipping managers with
necessary information at an acceptable level of accuracy to perform a variety of
managerial tasks (Mishra and Vaysman, 2001). The variable costing system collects
information on merely cost items that change proportionally with the number of
products/services produced and entirely ignores the fixed or overhead cost not directly
related to products/services produced. Like the variable costing system, the absorption
costing system track all variable costs but also assumes that all overhead costs are directly
related to the level of products/services produced (Maskell, 2006). Therefore, it allocates
overhead costs, usually accumulated at department level, to products/services based on
volume-based drivers such as labour hours or labour costs (Al-Omiri and Drury, 2007).
The mechanism followed by the TAS could be acceptable in an operations environment
where overhead costs constitute a small proportion of the overall operations costs and a
relatively small number of standardised products/services are produced (Lamminmaki and
Drury, 2001; Brignall et al., 1991, Datar et al., 1991; Cooper and Kaplan, 1988).
However, the old operational environment in which such systems have been argued to be
valid has changed (Drury and Tayles, 2005; Brignall, 1997). In the new environment, (i)
the growing reliance on automation has increased substantially the proportion of overhead
costs by reducing the labour content of operations; (ii) more customised products/services
have been produced which has further increased the overhead costs (Lamminmaki and
Drury, 2001; Brignall et al., 1991). Some researchers believe that these changes have
brought the TAS to its end (Cooper and Kaplan, 1992). TAS, in the new operational
environment, is argued to generate distorted cost information because of its reliance on
volume-based cost drivers for allocating the increasing overhead costs which does not
reflect the true resources consumed by products/services (Lamminmaki and Drury, 2001;
Cooper and Kaplan, 1992; Datar et al., 1991).
55
By either totally ignoring the increasing overhead costs or merely focusing on absorbing
all overhead costs (Khataie and Bulgak, 2013; Chiarini, 2012; Mishra and Vaysman,
2001), TAS is believed to generate aggregate and misleading information which keeps
waste hidden in the overhead allocation rate and does not reveal areas for improvements
(Khataie and Bulgak, 2013, Maskell, 2006; Kennedy and Affleck-Graves, 2001; Chenhall
and Langfield-Smith, 1998; Toomey, 1994). According to Datar et al. (1991), TAS may
stifle process improvement initiatives by providing inaccurate cost information to decision
makers or sending incorrect signals in relation to the effectiveness of a process innovation
which lead decision makers to mistakenly cease such an innovation. For example, this can
happen when TAS through its distorted cost information reveals that a product/service A
is profitable when in fact it is not while a product/service B is not profitable when in fact it
is. In such scenario, a company may not attempt to improve the process of producing
product/service A because it is seen to be profitable while the company may decide to
apply lean practices, or others, to improve the process of producing product/service B. In
this case, the need to improve the process of product/service A is concealed, while the
attempt to improve product/service B may not result in a substantial improvement leading
to questioning the effectiveness of the improvement program implemented. Both Datar et
al. (1991) and Cooper and Maskell (2008) report examples of companies who were about
to cease the implementation of lean practices mainly because of the misleading
information of the TAS used. As a result, a new accounting system, called activity-based
costing system (ABC), has been suggested as an alternative (Ruiz-de-Arbulo-Lopez et al.,
2013; Abu Mansor et al., 2012; Grasso, 2005; Kennedy and Affleck-Graves, 2001; Cooper
and Kaplan, 1992).
ABC improves the overhead cost allocation process in two different ways (Mishra and
Vaysman, 2001). Unlike TAS, ABC (i) breaks down processes into activities and uses
these activities to accumulate overhead cost, and (ii) relies not only on volume-based but
also non-volume-based cost drivers to allocate overhead cost to final cost objects (e.g.
products/services) (Banker et al., 2008; Datar et al., 1991). By doing so, ABC is thought to
generate more accurate and detailed cost information at activity level and measure the true
56
levels of resources consumed by products/services (Khataie and Bulgak, 2013; Tsai and
Lai, 2007; Cagwin and Bouwman, 2002; Clarke and Mullins, 2001; Mishra and Vaysman,
2001; Ittner, 1999; Cooper and Kaplan, 1992). By focusing on activities and their cost
drivers, ABC seems to be a valuable method which supports lean system by (a)
highlighting value-adding activities and non-value adding activities (NVA), (b) sending a
correct signal to decision makers on processes which truly require improvements, and (c)
helping lean companies prioritise their improvement efforts (Khataie and Bulgak, 2013;
Banker et al., 2008; Larson and Kerr, 2007; Grasso, 2005; Ittner, 1999; Datar et al., 1991;
Turney and Anderson, 1989).
Supporting the above argument on the relationship between ABC and lean system, Innes
and Mitchell (1995) surveyed the largest UK companies and found them to depend on the
measures generated from ABC to support other improvement initiatives such as
continuous improvement, TQM and JIT considered essential parts of lean system. Adam
(1996) argued that when ABC is used in conjunction with other process improvement
systems such as quality and lean system, companies enjoy a higher level of benefits.
Khataie and Bulgak (2013) used a system dynamics modelling tool and revealed the
importance of ABC in achieving the aim of lean system. Another simulation study by Li et
al. (2012) also showed that ABC has superiority over TAS in bridging the gap between
operational and financial improvements of lean companies. Chiarini (2012) also
demonstrated the advantages of ABC in a medium-sized lean firm. Banker et al. (2008)
reported empirical evidence indicating that manufacturing companies adopting ABC are
more likely to adopt lean practices in their operations. Abu Mansor et al. (2012) explored
the usefulness of ABC information for decision making in a telecommunication company.
Using data from 181 ABC users, Abu Mansor et al. (2012) found that ABC is critical for
improving areas of budgeting, planning and uncovering avenues for improvement in other
business areas which can be targeted by lean practices.
2.6.2 Business strategy and lean system
Business strategy usually expresses how a company chooses to compete against its
competitors and attain a competitive advantage in the market (Bruggeman and Stede,
57
1993; Porter, 1980). Porter (1980) classifies business strategy into differentiation and cost
leadership, while Miles and Snow (1978) differentiate between prospectors, analysers and
defenders. These two taxonomies, as proposed in the strategic management literature,
overlap in that differentiation/prospectors can be at one end of a continuum while cost
leadership/defenders can be at the other end (Chenhall, 2003; Langfield-Smith, 1997).
However, the inventors of those taxonomies argue that different organisational structure,
accounting systems and priorities are required for the different strategies at the two ends
of the continuum (Porter, 1980; Miles and Snow, 1978).
For instance, cost leaders/defenders are usually characterised with low variety and
standardised products/services and operate in a relatively stable market environment
(Ward et al., 2007; Gosselin, 1997; Lei et al., 1996; Ward et al., 1996; Bruggeman and
Stede, 1993). They focus heavily on controlling and reducing cost especially in areas such
as research and development and advertising (Frey and Gordon, 1999, Porter, 1980; Miles
and Snow, 1978). In contrast, differentiators/prospectors operate in a relatively more
volatile environment and compete through product/service innovation and market
development (Gosselin, 1997; Bruggeman and Stede, 1993). They seek developing a
wider range of more customised products/services with high focus on quality while they
strive for flexibility in responding to changing customers’ needs (Kennedy and Widener,
2008; Ward et al., 2007; Lei et al., 1996; Ward et al., 1996).
The relationship between business strategy and operations strategy has been studied by
several researchers in the operations and strategic management literature (e.g. Ward et al.,
2007; Lei et al., 1996; Ward et al., 1996). The core argument is that operations strategy
should work in line with the objectives of the overall business strategy (Joshi et al., 2003).
That is, business strategy is expected to shape the operations strategy to achieve firm
specific objectives (Amoako-Gyampah and Acquaah, 2008; Ward and Duray, 2000). As a
result, it is logical to expect an influence from the business strategy on the implementation
of lean practices chosen as an operations strategy (Ward et al., 2007; Baines and
Langfield-Smith, 2003; Chenhall and Langfield-Smith, 1998).
58
Empirical examinations of this premise in the service context are lacking and those in the
manufacturing context provide mixed results. While Qi et al. (2011) have found cost
leadership strategy to be more compatible with lean system than the differentiation
strategy, others have demonstrated the reverse (Ward et al., 2007; Chenhall and Langfield-
Smith, 1998).
2.6.3 Firm age and lean system
Firm age can influence the adoption of lean practices in different ways (González-Benito,
2005; Shah and Ward, 2003). On the one hand, old firms are more likely to be more
experienced in running businesses in comparison with young firms (Coad et al., 2013;
Glancey, 1998). This accumulated knowledge and experience may help old firms to be
more efficient than less experienced firms and prevents the need for adopting lean
practices to improve efficiency (Coad et al., 2013; Lundvall and Battese, 2000; Glancey,
1998). In addition, old firms may suffer from rigidity and inflexibility in responding to
market changes and adopting new innovations such as lean practices (Coad et al., 2013;
Wagner et al., 2012; González-Benito, 2005; Shah and Ward, 2003). However, while Shah
and Ward (2003) found empirical evidence regarding the negative impact of firm age on
the adoption of some lean practices (e.g. cross functional work force, cycle time reduction,
JIT/continuous flow production, maintenance optimization, reengineered production
process and self-directed work teams), they also found a positive relation between firm
age and other lean practices (e.g. planning and scheduling strategies, safety improvement
programs, and total quality management programs). Therefore, the authors concluded that
the effect of firm age is not always in the predicted direction although it should be taken
into account.
2.6.4 Firm size and lean system
Firm size can also have an impact on the adoption of lean practices. Large firms are
argued to have more financial and human resources. The higher level of resources
available for large firms can be advantageous by allowing for more experimentation with
new technologies and innovations (e.g. lean practices) that may improve their productivity
59
and efficiency (Coad et al., 2013; Wagner et al., 2012; Galende and de la Fuente, 2003;
Shah and Ward, 2003). In contrast, large firms have more complex operations
administrative tasks and therefore can be more reluctant or slower in adopting innovative
methods and techniques that are capable of improving their performance (Shah and Ward,
2003; Hannan and Freeman, 1984). In their empirical study, Shah and Ward (2003) found
evidence of a positive association between firm size and 20 out of 22 lean practices. This
highlights the importance of taking into account the effect of firm size when studying the
lean-performance association.
2.6.5 Evaluation of the lean-context literature
Although the literature reviewed in the subsections 2.6.1 and 2.6.2 has been useful in
providing information on the role of ABC and business strategy in the lean context, it still
has some limitations that deserve to be considered. First, most studies which highlighted
the shortcomings of the TAS and the advantages of ABC in the lean context were either
conceptual or case-based studies. As a result, there has been very little empirical research
devoted to verifying the anecdotal evidence suggested by those studies (Fullerton et al.,
2013; Li et al., 2012; Banker et al., 2008). Empirical examinations of the possible effect of
ABC on lean service are critical for the ABC literature as it offers an alternative view on
how ABC can impact on firm performance. While the majority of researchers examined
the direct impact of ABC on performance (e.g. Sheu and Pan, 2009; Cagwin and
Bouwman, 2002; Ittner et al., 2002; Kennedy and Affleck-Graves, 2001), little attention
has been paid to its potential indirect impact on performance acting through lean service
(Banker et al., 2008).
Second, studies reviewed in the two subsections were distinctive in that they focused on
either the lean-accounting association or the strategy-lean relationship. With the exception
of the study of Baines and Langfield-Smith (2003), neither of the studies has attempted to
integrate the lean-accounting literature with the strategy-lean literature despite the
significant insights that can be obtained from that. The integration of the two streams of
literature allows for taking into account the impact of business strategy on ABC
documented by several researchers (e.g. Hammad et al., 2010; Baines and Langfield-
60
Smith, 2003; Chenhall and Langfield-Smith, 1998; Gosselin, 1997), which helps to
uncover the intervening role of ABC in the lean-strategy association. Consequently, the
impact of business strategy on lean service can be decomposed into its direct and indirect
elements, which offers a better understanding of such impact (Luft and Shields, 2003;
Shields et al., 2000). In addition, the inclusion of business strategy when examining the
effect of ABC on lean service is essential to avoid endogeneity issues caused by correlated
omitted variables and which bias the findings2 (Chenhall and Moers, 2007; Larcker and
Rusticus, 2007).
2.7 Conclusion and research gaps
In this chapter, the literature on lean system in manufacturing and services has been
reviewed. This included the provision of information on the evolution of lean system since
its introduction in 1950s. In addition, the various definitions of lean system along with its
structure and mechanism for improving firm performance have been highlighted.
Moreover, the literature concerning the possible role the accounting system and business
strategy in the lean context has been critically reviewed.
The literature on lean system in manufacturing was found to be more developed compared
to lean service. However, the empirical studies investigating the impact of lean
manufacturing on performance were limited in scope by their focus on either the impact of
individual practices or the additive effect of lean bundles on operational performance. The
likely non-additive effect of lean bundles suggested by Shah and Ward (2003) and the
effect on financial performance were far less investigated in this empirical literature.
The lean service literature was found to be growing with increasing number of
publications although they were mostly conceptual and case studies. Despite the growth in
the number of publications, the lean service literature was found to be strongly biased to
healthcare at the account of other perhaps equally important service industries such as
hotels, consultancy services, telecommunication industry, banking and financial service,
2 Endogeneity caused by correlated omitted variables occurs as a result of excluding one or more variables which are expected to
affect both the dependent and independent variable(s) in a model (Chenhall and Moers, 2007).
61
etc. Therefore, it is vital to supplement this growing body of literature with rigorous
empirical studies that focus on the neglected sectors while learning from the
manufacturing literature to overcome its limitations.
The focus on the lean-performance association directed the attention of researchers away
from the importance of shedding light on the impact of contextual variables on the
adoption of lean practices. Two distinctive streams of literature were found and reviewed
which highlighted the impact of ABC and business strategy on the implementation of lean
system. Although the current literature was clear in terms of the superiority of ABC over
TAS in the lean context, it was largely dominated by conceptual and case-based studies.
This fact necessitates rigorous empirical research to be conducted in order to verify the
anecdotal evidence on the positive role of ABC in the lean service. Further, investigating
the role of ABC and business strategy in the lean context in isolation was found to conceal
significant insights on their effect. Hence, a better understanding of their role requires the
integration of three different streams of literature, namely, the accounting-lean literature,
the strategy-lean literature and the strategy-accounting literature.
In short, rigorous empirical research, that (a) covers a wide range of service industries to
provide more generalisable results, (b) brings to light not only the additive but also the
non-additive impact of lean bundles on both operational and financial performance, and
(c) examines simultaneously the impact of ABC and business strategy on lean service can
be of significant importance (Arlbjørn and Freytag, 2013; Fullerton et al., 2013;
Malmbrandt and Åhlström, 2013; Suárez-Barraza et al., 2012; Holm and Ahlstrom, 2010b;
Banker et al., 2008; Shah and Ward, 2003).
62
Chapter 3 : Theoretical framework and hypotheses
development
3.1 Introduction
In the previous chapter, the shortcomings characterising the existing literature of lean
system in manufacturing and services were exposed. In addition, limitations of the streams
of literature focusing on the role of ABC and business strategy in the lean context were
also highlighted. This chapter aims to address these limitations by developing a theoretical
model that brings to light the full potential of lean service and clarifies the impact of ABC
and business strategy on lean service. The remainder of this chapter is organised as
follows. In the next section, the constructs forming the theoretical model developed in this
study are identified and explicitly reported. In the third section, the two core theories
guiding the development of the theoretical model are presented. Section four focuses on
describing and explaining the theoretical model while section five articulates the
hypotheses linking together the constructs of the theoretical model. The last section of this
chapter provides a summary of information presented in this chapter.
3.2 Constructs of the theoretical model
3.2.1 Identification of the relevant literature
Given the service nature of this research, the first step in determining the constructs of this
study is to define the term “service” and identify the relevant literature accordingly. To
distinguish publications on lean system in services from those discussing lean system in
non-service context, the following definition of services has been adopted: service firms
are any firm which is not involved in manufacturing, agriculture, mining and construction
industries. This residual perspective in defining services has been criticised by Sampson
and Froehle (2006) who proposed an alternative way called the Unified System Theory to
differentiate between services and non-services. However, the new method introduced by
Sampson and Froehle (2006) differentiates between services and non-services at a process
level. Consequently, the use of the Unified System Theory is less applicable to this
63
research given that the unit of analysis here is the firm rather than the process. In addition,
the definition adopted in this study has been adopted by a large number of researchers
(e.g. Zaman et al., 2013; Abdelaziz et al., 2011; Saidur et al., 2011; Christie et al., 2003).
Keeping in mind this definition, a systematic search for lean service publications began by
surveying publications in five well-known databases using key words including “lean”,
“process improvement”, “system thinking” and “more with less”. Those databases
included: Business Source Premier, ABI/INFORM Research, Emerald, Science Direct and
Scopus. All articles reporting any of the aforementioned key words in the title, abstract or
key words were collected for further examinations. The title and abstract of each article
were examined to determine those articles on lean service. After identifying publications
on lean service, references listed at the end of each article were traced to collect all
possible relevant articles. Through this process, 221 articles have been found (up to the
end of May 2013) and presented in chapter 2. In addition, these articles have been used to
extract required information, i.e., lean practices and outcome of their implementation for
constructing the model.
3.2.2 Identification of lean service practices
The identification of lean practices is not a trivial task given the confusion surrounding the
concept (Lewis, 2000) (see section 2.2). For instance, the human-based practices are
emphasised and argued to be crucial for any improvement system where lean is not an
exception (Höök and Stehn, 2008). Based on that, some authors include explicitly or
implicitly the human-based practices such as education/training, employees’ involvement
and empowerment, multi-skilled/multi-function employees and teamwork in the lean
toolbox (e.g. Staats et al., 2011; Holden, 2010; Kuriger et al., 2010; Poksinska, 2010;
Manos et al., 2006). Shah and Ward (2003) involve HRM practices in the lean toolbox to
conclude that lean system comprises four bundles namely, JIT, TQM, TPM and HRM
practices, each of which has its own items. In contrast, other researchers adopt a different
perspective believing that HRM practices are important to a successful lean
implementation, and consequently they are a prerequisite for lean system (e.g. Suarez-
Barraza and Ramis-Pujol, 2010; Ehrlich, 2006; Comm and Mathaisel, 2005a). Pettersen
64
(2009) reports that the findings of his study contradict those of Shah and Ward (2003). He
finds that HRM is not a basic characteristic of lean although it is important to present.
Pont et al. (2008) consider HRM practices as one of the lean bundles and report the
importance of their implementation first in the lean journey. That, however, implicitly
supports the need to differentiate between HRM and other lean bundles. Fullerton and
Wempe (2009) separate employees’ involvement from other lean tools namely, cellular
manufacturing and quality improvement. Finally, studying the impact of JIT practices on
plants performance, Sakakibara et al. (1997) and Ahmad et al. (2003) perceive similar
practices as supportive and infrastructure practices necessary for an effective JIT system.
Consequently, in this study, I follow Shah and Ward (2007) in viewing lean service as a
socio-technical system that has two distinctive sets of practices. HRM practices, however,
discussed above and other practices identified in the literature will represent the social side
of the system.
Changing the facility layout (CFL) is another controversial point. For instance, some
researchers view CFL as a requirement for moving away from a department-based
organisation to a process-based organisation (Yasin et al., 2003) which is needed for the
group technology concept. In contrast, others consider CFL to be one of the lean
techniques that could be employed to attack one or more of waste elements (Holden, 2010;
Holm and Ahlstrom, 2010a; Poksinska, 2010; Manos et al., 2006; Tonya, 2004; Allway
and Corbett, 2002). Theoretically and regardless of leading to a complete process-based
layout, the layout of an organisation can be modified so that any unnecessary movements
of employees and/or inefficient use of space can be eliminated (Hameri, 2010). Therefore,
CFL will be included in the lean technical practices (LTPs) that an organisation can use to
eliminate waste.
This differentiation between LTPs and lean social practices (LSPs) is highly important
given that some companies may not adopt all practices (technical and social). Therefore, if
the aforementioned LSPs are combined with LTPs in one comprehensive group, they may
not be considered for implementation leading to a less successful overall outcome and
consequently distorting the reputation of lean service. In addition, classifying lean service
65
practices into technical and social practices allows for applying and empirically testing the
mechanism of STS explicated in section 3.3. Bearing in mind the above discussion, the 54
lean service practices identified through carefully reading the 221 publications have been
classified into 37 LTPs and 17 LSPs as presented in Tables 3-1 and 3-2 with their
supportive references. These two sets of practices represent the source of the two
constructs of the theoretical model.
Table 3-1: Lean technical practices (*
, ^)
No. Practice References
1 5Ss Ehrlich(2006),Holden(2010),Poksinska (2010), Arlbjørn et al. (2011), Burgess and
Radnor(2010), Manos et al. (2006), Fillingham (2007), Esain et al. (2008), Emiliani (2004),
Bushell et al. (2002), Suarez-Barraza et al. (2009), Wayne (2005), Brewton (2009), Tiplady
(2010), Finigan and Humphries (2006), Maguad (2007), Julien and Tjahjono (2009), Haque
And James-Moore (2004), Keen (2011), Pedersen and Huniche (2011), Wenchao Song et al.
(2009), Kaplan and Patterson (2008), Markovitz (2012), Chadha et al. (2012), Radnor et al.
(2012), Schulze and Störmer (2012)
2 A3 report Holden (2010), Jimmerson et al. (2005), Doman (2011), Qudrat-Ullah et al. (2012)
3 Automation Holden (2010), Poksinska (2010), Bortolotti and Romano (2010), Manos et al. (2006),
Ahluwalia et al. (2004), Lodge and Bamford (2008), Wayne (2005), Julien and Tjahjono
(2009), Åhlström (2004), Carter et al. (2011), Wenchao Song et al. (2009), Bortolotti and
Romano (2012)
4 Change management Manos et al. (2006)
5 Continuous improvement Dickson et al. (2009), Ehrlich (2006), Poksinska (2010), Piercy and Rich (2009a), Manos et
al. (2006), Emiliani (2004), Alagaraja (2010), Maguad (2007), Kuriger et al. (2010), Yavas
and Yasin (2001), Hagan (2011), Qudrat-Ullah et al. (2012)
6 Eliminating loop-backs Swank (2003)
7 Group technology Piercy and Rich (2009b), Ehrlich (2006), Holden (2010), Swank (2003), Arlbjørn et al.
(2011), Nielsen and Edwards (2010), Burgess and Radnor (2010), Manos et al. (2006),
Arbos (2002), Ben-Tovim et al. (2007), Alagaraja (2010), Hyer and Wemmerlöv (2002),
Tatikonda (2007), Cuatrecasas (2004), Middleton et al. (2005)
8 Changing the facility
layout
Allway and Corbett (2002), Holden (2010), Poksinska (2010), Holm and Ahlstrom (2010),
Manos et al. (2006), Tonya (2004), Cuatrecasas (2004), Nelson-Peterson and Leppa (2007)
9 Just in Time Cooper and Mohabeersingh (2008), Holden (2010), Poksinska (2010), Arlbjørn et al.
(2011), Manos et al. (2006), Emiliani (2004), Alagaraja (2010), Åhlström (2004),
Cuatrecasas (2004), Nelson-Peterson and Leppa (2007), Chadha et al. (2012)
10 Kaizen blitz Dickson et al. (2009), Holden (2010), Arlbjørn et al. (2011), Burgess and Radnor (2010),
Hines and Lethbridge (2008), Suarez-Barraza and Ramis-Pujol (2010), Suarez Barraza et al.
(2009), Kress (2008), Papadopoulos and Merali (2008), Montabon (2005), Graban and
Swartz (2012), Papadopoulos (2012), Radnor et al. (2012)
11 Kanban Holden (2010), Poksinska (2010), Arlbjørn et al. (2011), Manos et al. (2006), Alagaraja
(2010), Reinertsen (2005), Hagan (2011), Nelson-Peterson and Leppa (2007)
12 Mistakes proofing/Poka-
Yoke
Ehrlich (2006), Cooper and Mohabeersingh (2008), Holden (2010), Poksinska (2010),
Manos et al. (2006), Alagaraja (2010), Finigan and Humphries (2006), Maguad (2007),
Kuriger et al. (2010), Mirehei et al. (2011), Hagan (2011), Doman (2011)
13 Model cell, roll out Swank (2003), Graban and Swartz (2012)
14 Outsourcing Comm and Mathaisel (2005b)
15 Point of use storage Manos et al. (2006)
16 Policy
deployment/Hoshin
Poksinska (2010), Swank (2003), Emiliani (2004), Alagaraja (2010), Pejsa and Eng (2011),
Ball and Maleyeff (2003), Wayne (2005), Qudrat-Ullah et al. (2012)
66
Kanri
17 Process redesign Piercy and Rich (2009a), McQuade (2008), Suarez-Barraza and Ramis-Pujol (2010),Yavas
and Yasin (2001), Carter et al. (2011), Edwards et al. (2012), Chadha et al. (2012),
Bortolotti and Romano (2012)
18 Production
levelling/Heijunka
Poksinska (2010), Emiliani (2004), Staats et al. (2011), Pedersen and Huniche (2011)
19 Pull system Ehrlich (2006), Cooper and Mohabeersingh (2008), Poksinska (2010), Holm and Ahlstrom
(2010), Arlbjørn et al. (2011), Manos et al. (2006), Petersen and Wohlin (2010), Kuriger et
al. (2010), Reinertsen and Shaeffer (2005), Reinertsen (2005), Kress (2008), Mirehei et al.
(2011), Hagan (2011), Ball and Maleyeff (2003), Schulze and Störmer (2012)
20 Quality circles Swank (2003), Searcy (2009b)
21 Quality function
deployment
Emiliani (2004), Alagaraja (2010), Tatikonda (2007), Wang et al. (2012), Schulze and
Störmer (2012)
22 Quick set up time Arlbjørn et al. (2011), Manos et al. (2006), Arbos (2002), Finigan and Humphries (2006),
Maguad (2007)
23 Root cause analysis Ehrlich (2006), Holden (2010), Poksinska (2010), Jones et al. (1999), Petersen and Wohlin
(2010), Searcy (2009b), Villarreal et al. (2009), Haque And James-Moore (2004), Yavas
and Yasin (2001), Wang et al. (2012), Collar et al. (2012), Schulze and Störmer (2012)
24 Segregating complexity Holm and Ahlstrom (2010), Swank (2003), Nielsen and Edwards (2010), King et al. (2006)
25 Self inspection Manos et al. (2006), Maguad (2007)
26 Simplification Bortolotti and Romano (2010), Bortolotti and Romano (2012)
27 Single piece flow Poksinska (2010), Staats et al. (2011), Alagaraja (2010), Kuriger et al. (2010), Haque And
James-Moore (2004), Kress (2008), Mirehei et al. (2011), Nelson-Peterson and Leppa
(2007), Chadha et al. (2012), Bortolotti and Romano (2012)
28 Small lots Ehrlich (2006), Swank (2003), Manos et al. (2006), Arbos (2002), Brewton (2009), Kuriger
et al. (2010), Reinertsen and Shaeffer (2005), Reinertsen (2005), Kress (2008)
29 Standardisation Allway and Corbett (2002), Ehrlich (2006), Holden (2010), Poksinska (2010), Holm and
Ahlstrom (2010), Sprigg and Jackson (2006), Swank (2003), Bortolotti and Romano (2010),
Kosuge et al. (2010), Nielsen and Edwards (2010), Manos et al. (2006), Emiliani (2004),
Bushell et al. (2002), Staats et al. (2011), Alagaraja (2010), Haque And James-Moore
(2004), LaGanga (2011), Hagan (2011), Wenchao Song et al. (2009), Kaplan and Patterson
(2008), Nelson-Peterson and Leppa (2007), Middleton et al. (2005), Doman (2011),
Carlborg et al. (2013), Wang et al. (2012), Qudrat-Ullah et al. (2012), Chadha et al. (2012),
Bortolotti and Romano (2012), Jaca et al. (2012)
30 Takt time Allway and Corbett (2002), Poksinska (2010), Holm and Ahlstrom (2010), Swank (2003),
Arlbjørn et al. (2011), Arbos (2002), Emiliani (2004), Haque And James-Moore (2004),
Reinertsen (2005), Kress (2008), Cuatrecasas (2004), Middleton et al. (2005)
31 Total preventive
maintenance
Poksinska (2010), Arlbjørn et al. (2011), Manos et al. (2006), Arbos (2002), Emiliani
(2004), Finigan and Humphries (2006), Maguad (2007), Åhlström (2004)
32 Total quality Kuriger et al. (2010), Mirehei et al. (2011)
33 Use of new technologies Jones et al. (1999), Hines and Lethbridge (2008), Comm and Mathaisel (2005b), Tischler
(2006)
34 Value stream mapping Dickson et al. (2009), Piercy and Rich (2009b), Ehrlich (2006), Holden (2010), Poksinska
(2010), Arlbjørn et al. (2011), Bortolotti and Romano (2010), Piercy and Rich (2009a),
Nielsen and Edwards (2010), Burgess and Radnor (2010), King et al. (2006), Jimmerson et
al. (2005), Ahluwalia et al. (2004), Fillingham (2007), Jones et al. (1999), Lodge and
Bamford (2008), Bushell et al. (2002), McQuade (2008), Suarez-Barraza and Ramis-Pujol
(2010), Hines et al. (2008), Ben-Tovim et al. (2007), Suarez Barraza et al. (2009), Staats et
al. (2011), Alagaraja (2010), Tonya (2004), Wayne (2005), Searcy (2009b), Tiplady (2010),
Maguad (2007), Villarreal et al. (2009), Julien and Tjahjono (2009), Haque And James-
Moore (2004), Keen (2011), Kress (2008), LaGanga (2011), Pedersen and Huniche (2011),
Papadopoulos and Merali (2008), Wenchao Song et al. (2009), Tischler (2006), Chaneski
(2005), Doman (2011), Wang et al. (2012), Chadha et al. (2012), Bortolotti and Romano
(2012), Schulze and Störmer (2012), Vlachos and Bogdanovic (2013)
35 Vertical information
system
Holm and Ahlstrom (2010), Åhlström (2004)
67
36 Visualisation Holden (2010), Poksinska (2010), Arlbjørn et al. (2011), Manos et al. (2006), Fillingham
(2007), Emiliani (2004), Bushell et al. (2002), Staats et al. (2011), Alagaraja (2010), Wayne
(2005), Brewton (2009), Finigan and Humphries (2006), Haque And James-Moore (2004),
Keen (2011), Wenchao Song et al. (2009), Tischler (2006), Kaplan and Patterson (2008),
Nelson-Peterson and Leppa (2007)
37 Work load balancing Swank (2003), Brewton (2009), Kuriger et al. (2010), Cuatrecasas (2004), Mirehei et al.
(2011), Wenchao Song et al. (2009), Middleton et al. (2005)
* All practices printed in bold will not be included in the questionnaire because they were not reported by at least five
studies (see subsection 4.10.1)
^ All practices printed in italics violated the normality assumption and therefore they will not be included in the factor
analysis and further empirical examination (see subsection 5.2.2)
Table 3-2: Lean social practices*
No. Social practices References
1 An appropriate rewarding system Piercy and Rich (2009b), Ehrlich (2006), Holden (2010), Wayne (2005), Jaca et al.
(2012)
2 Customer involvement Holm and Ahlstrom (2010), Suarez-Barraza and Ramis-Pujol (2010)
3 Effective Communication System Allway and Corbett (2002), Holden (2010), Swank (2003), Manos et al. (2006),
Hines and Lethbridge (2008), Suarez-Barraza and Ramis-Pujol (2010), Hines et al.
(2008), Comm and Mathaisel (2005a), Pejsa and Eng (2011), Jaca et al. (2012)
4 Employee empowerment Holden (2010), Jones et al. (1999), Comm and Mathaisel (2005a), Graban and
Swartz (2012), deHaan et al. (2012), Bortolotti and Romano (2012), Collar et al.
(2012)
5 Employees commitment Dickson et al. (2009), Poksinska (2010), Carter et al. (2011), Bortolotti and
Romano (2012), Schulze and Störmer (2012)
6 Employees involvement Piercy and Rich (2009b), Ehrlich (2006), Holden (2010), Swank (2003), Bortolotti
and Romano (2010), Manos et al. (2006), Suarez-Barraza and Ramis-Pujol (2010),
Hines et al. (2008), Tonya (2004), Julien and Tjahjono (2009), Kress (2008),
Graban and Swartz (2012), deHaan et al. (2012), Bortolotti and Romano (2012),
Collar et al. (2012), Schulze and Störmer (2012), Jaca et al. (2012)
7 Establishing a long-term
relation with suppliers
Swank (2003), Wang et al. (2012), Qudrat-Ullah et al. (2012)
8 Establishing environment for
change
Comm and Mathaisel (2005a), Graban and Swartz (2012)
9 Having multifunctional
employees
Dickson et al. (2009), Ehrlich (2006), Arbos (2002), Tonya (2004), Moayed and
Shell (2009), Cuatrecasas (2004), LaGanga (2011), Chadha et al. (2012)
10 Improving teamwork spirit Suarez-Barraza and Ramis-Pujol (2010), Graban and Swartz (2012), Jaca et al.
(2012)
11 Leadership Allway and Corbett (2002), Swank (2003), Suarez-Barraza and Ramis-Pujol
(2010), Comm and Mathaisel (2005a), Jaaron and Backhouse (2011), Keen (2011),
Qudrat-Ullah et al. (2012), Schulze and Störmer (2012)
12 Modifying the terminology to
suit services
Hines et al. (2008)
13 Obtaining management support Dickson et al. (2009), Allway and Corbett (2002), Piercy and Rich (2009b),
Poksinska (2010), Holm and Ahlstrom (2010), Swank (2003), Bortolotti and
Romano (2010), Piercy and Rich (2009a), Burgess and Radnor (2010), King et al.
(2006), Jimmerson et al. (2005), Hines and Lethbridge (2008), Suarez-Barraza and
Ramis-Pujol (2010), Tischler (2006), Towne (2006), Graban and Swartz (2012),
Papadopoulos (2012), Jaca et al. (2012)
14 Performance measurement
system
Piercy and Rich (2009b), Ehrlich (2006), Swank (2003), Burgess and Radnor
(2010), Suarez-Barraza and Ramis-Pujol (2010), Bhasin (2008), Comm and
Mathaisel (2005a), Comm and Mathaisel (2005b), Kennedy et al. (2007), Kress
(2008), Bortolotti and Romano (2012)
15 Posting performance results Swank (2003), Middleton et al. (2005)
68
16 Providing justifications for
implementing the practices
Jaaron and Backhouse (2011)
17 Training Dickson et al. (2009), Piercy and Rich (2009b), Ehrlich (2006), Holden (2010),
Poksinska (2010), Holm and Ahlstrom (2010), Piercy and Rich (2009a), Burgess
and Radnor (2010), Manos et al. (2006), King et al. (2006), Jimmerson et al.
(2005), Suarez-Barraza and Ramis-Pujol (2010), Hines et al. (2008), Staats et al.
(2011), Comm and Mathaisel (2005a), Comm and Mathaisel (2005b), Tonya
(2004), Wayne (2005), Searcy (2009b), Keen (2011), Kress (2008), Cuatrecasas
(2004), Mirehei et al. (2011), Carter et al. (2011), Tischler (2006), Graban and
Swartz (2012), Schulze and Störmer (2012), Jaca et al. (2012)
* All practices printed in bold will not be included in the questionnaire because they were not reported by at least five studies
(see subsection 4.10.1)
3.2.3 The performance outcome of lean service practices
The third critical construct in the theoretical model developed in this study is the benefits
likely to result from the implementation of lean service practices. The systematic review
of the existing lean service literature indicates the presence of 20 benefits that can be
attributed to lean service practices as presented in Table 3-3 with associated references.
The 20 benefits listed in Table 3-3 are comprehensive in that they represent not only
operational performance of service firms but also their financial performance. Therefore,
the items presented in Table 3-3 can capture the multi-dimensional nature of
organisational performance to overcome limitations of the majority of previous empirical
lean-performance studies which focused only on one dimension of organisational
performance, namely the operational performance (as highlighted in the introduction
chapter and subsection 2.4.4).
Table 3-3: Benefits of lean service*
N0. Benefits References
1 Freeing staff time Piercy and Rich (2009b), Ehrlich (2006), Jimmerson et al. (2005), Searcy (2009), Hagan
(2011), Papadopoulos (2012), Markovitz (2012), Bortolotti and Romano (2012)
2 Identification and elimination
of waste
Ehrlich (2006), Swank (2003), Hines and Lethbridge (2008), McQuade (2008), Comm
and Mathaisel (2005a), Maguad (2007), Julien and Tjahjono (2009), Kaplan and
Patterson (2008), Nelson-Peterson and Leppa (2007), Chadha et al. (2012), Bortolotti
and Romano (2012), Collar et al. (2012), Schulze and Störmer (2012)
3 Improvement in capacity Holden (2010), Poksinska (2010), Nielsen and Edwards (2010), Burgess and Radnor
(2010), King et al. (2006), Ben-Tovim et al. (2007), LaGanga (2011), Hagan (2011),
Chadha et al. (2012)
4 Improvement in customer
perception of product/service
quality
Piercy and Rich (2009b), Ehrlich (2006), Bortolotti and Romano (2010), Arbos (2002),
Hyer and Wemmerlöv (2002), Hagan (2011), Nelson-Peterson and Leppa (2007)
5 Improvement in customer
satisfaction
Piercy and Rich (2009b), Ehrlich (2006), Poksinska (2010), Bortolotti and Romano
(2010), Piercy and Rich (2009a), Jimmerson et al. (2005), Emiliani (2004), Hines and
69
Lethbridge (2008), Suarez-Barraza and Ramis-Pujol (2010), Hines et al. (2008), Yavas
and Yasin (2001), Pejsa and Eng (2011), Kaplan and Patterson (2008), Carlborg et al.
(2013), Edwards et al. (2012), Bortolotti and Romano (2012)
6 Improvement in employees
satisfaction and their
performance
Piercy and Rich (2009b), Poksinska (2010), Swank (2003), Nielsen and Edwards (2010),
Burgess and Radnor (2010), Jimmerson et al. (2005), Fillingham (2007), Hines et al.
(2008), Pejsa and Eng (2011), Kaplan and Patterson (2008), deHaan et al. (2012),
Edwards et al. (2012), Jaca et al. (2012)
7 Improvement in employees
understanding of the process
Swank (2003), Burgess and Radnor (2010), Esain et al. (2008), Bushell et al. (2002),
Maguad (2007), Radnor et al. (2012)
8 Improvement in operational
efficiency
Cooper and Mohabeersingh (2008a), Bortolotti and Romano (2010), Cooper and
Mohabeersingh (2008b), Malladi et al. (2010), Comm and Mathaisel (2005a), Carlborg
et al. (2013), Collar et al. (2012)
9 Improvement in process
flexibility
Bortolotti and Romano (2010), Kosuge et al. (2010), Chadha et al. (2012)
10 Improvement in productivity Allway and Corbett (2002), Bortolotti and Romano (2010), Arbos (2002), Bhatia and
Drew (2007), Staats et al. (2011), Pejsa and Eng (2011), Carlborg et al. (2013),
Bortolotti and Romano (2012), Jaca et al. (2012)
11 Improvement in the
organisation of work areas
Poksinska (2010), Manos et al. (2006), Suarez Barraza et al. (2009), Radnor et al. (2012)
12 Reduction in costs Piercy and Rich (2009b), Ehrlich (2006), Cooper and Mohabeersingh (2008b),
Poksinska (2010), Swank (2003), Bortolotti and Romano (2010), Piercy and Rich
(2009a), Nielsen and Edwards (2010), Jimmerson et al. (2005), Jones et al. (1999),
Arbos (2002), McQuade (2008), Ben-Tovim et al. (2007), Cooper and Mohabeersingh
(2008a), Bhatia and Drew (2007), Malladi et al. (2010), Villarreal et al. (2009), Julien
and Tjahjono (2009), Hagan (2011), Kaplan and Patterson (2008), Bortolotti and
Romano (2012)
13 Reduction in inventory Poksinska (2010), Bortolotti and Romano (2010), Manos et al. (2006), Jones et al.
(1999), Kaplan and Patterson (2008)
14 Reduction in lead time and
cycle time
Piercy and Rich (2009b), Ehrlich (2006), Cooper and Mohabeersingh (2008b), Holden
(2010), Poksinska (2010), Swank (2003), Bortolotti and Romano (2010), Piercy and
Rich (2009a), Nielsen and Edwards (2010), King et al. (2006), Fillingham (2007), Arbos
(2002), Lodge and Bamford (2008), McQuade (2008), Suarez-Barraza and Ramis-Pujol
(2010), Hines et al. (2008), Ben-Tovim et al. (2007), Cooper and Mohabeersingh
(2008a), Bhatia and Drew (2007), Suarez Barraza et al. (2009), Staats et al. (2011), Hyer
and Wemmerlöv (2002), Yavas and Yasin (2001), Hagan (2011), Tischler (2006),
Nelson-Peterson and Leppa (2007), Papadopoulos (2012), Edwards et al. (2012), Radnor
et al. (2012)
15 Reduction in reworks Jones et al. (1999), Hyer and Wemmerlöv (2002)
16 Reduction in staff turnover
and absenteeism
Piercy and Rich (2009b), deHaan et al. (2012)
17 Reduction in the number of
human errors
Poksinska (2010), Swank (2003), Jimmerson et al. (2005), Searcy (2009), Hyer and
Wemmerlöv (2002), Hagan (2011)
18 Reduction in work in
process
Piercy and Rich (2009b), Ehrlich (2006), Swank (2003)
19 Savings in space Nielsen and Edwards (2010), Manos et al. (2006), Suarez Barraza et al. (2009)
20 Profitability Allway and Corbett (2002), Ehrlich (2006), Bhatia and Drew (2007), Pejsa and Eng
(2011)
* All expected benefits printed in bold will not be included in the questionnaire because they were not reported by at least five
researchers (see subsection 4.10.2)
70
3.2.4 Contextual variables
To spot potential contextual variables which may affect either firm performance or the
adoption of lean practices or both, the service literature in the field of management
accounting, business strategy, strategic management and the research focusing on the
performance of service firms in general has been reviewed. However, given the scarcity of
such literature, I followed Auzair and Langfield-Smith (2005) by including relevant
research from the manufacturing literature. Auzair and Langfield-Smith (2005) pointed
out that when focusing on for-profit service firms as the case of this research, the literature
on manufacturing companies can be relied on as these types of companies share the same
goal which is profit maximisation. The thorough search for likely influential contextual
variables has resulted in a list of four contextual variables, namely firm size, firm age,
management accounting system (represented by ABC in this study) and business strategy.
More in-depth information about the role of each contextual variable is provided in section
5 of this chapter.
3.3 The theoretical foundation of the research
3.3.1 Socio-technical systems theory (STS)
The socio-technical system (STS) was developed at Tavistock Institute of Human
Relations in London and was reported on through a series of research papers written by
Eric Trist and his colleagues (e.g. Trist, 1981; Emery and Trist 1965; Trist and Bamforth,
1951). This theory underlines the notion that organisations consist of two separate, but
interdependent, systems: a technical system and a social system (Appelbaum, 1997; Fox,
1995; Trist, 1981). The technical system comprises equipments, tools, techniques and
processes, while the social system comprises people and relationships among them (Trist,
1981; Trist and Bamforth, 1951). The argument is that higher reliance on the technical
system as a response to changes in the environment can be more effective if it is
accompanied by a corresponding increase in the social system (Fox, 1995; Huber and
Brown, 1991; Trist and Bamforth, 1951). Therefore, although each system can be
described as a stand-alone system, the optimal performance of an organisation can only be
obtained by the joint optimisation of both systems (Zu, 2009; Manz and Stewart, 1997;
71
Fox, 1995; Trist, 1981; Emery and Trist, 1965). In other words, the sole focus on one
system at the expense of the other prevents adopters from enjoying a higher level of
benefits that would have been realised if they would have focused on the other
complementary system.
Given the notion of STS highlighted above, its use in the operations management
literature is evidently growing whether in manufacturing or services (Baxter and
Sommerville, 2011; Manz and Stewart, 1997). For instance, Manz and Stewart (1997)
provide a theoretical model that presents TQM from the STS perspective and report five
propositions where a synergy between the technical and social systems of TQM has been
expected. Similarly, Huber and Brown (1991) conduct a critical theoretical work to
convey an argument in which they highlight the importance of changes in the social
practices as a result of adopting the cellular manufacturing concept for the latter to be
more effective. In more recent work, Zu (2009) attempts to explain, using STS
perspective, the controversial results reported on the effectiveness of quality management
technical practices and quality management infrastructural practices. The author uses data
from 226 manufacturing firms and finds that the infrastructural practices are strongly
supportive to the technical practices in improving quality performance. In this line of
empirical research adopting the STS perspective, the very few studies including those of
Dabhilkar and Åhlström (2013) and Das and Jayaram (2007) found in the lean literature
and explained earlier in subsection 2.4.3 can be included.
The importance of the STS to the current research stems from its use to (i) classify the
long list of lean practices into LTPs and LSPs (see Tables 3-1 and 3-2), and (ii) to support
the theoretical argument pointing to the expected synergy between the two sets of
practices.
3.3.2 Contingency theory
The contingency theory (CT) is a critical approach that has contributed significantly to
different research fields including operations management (e.g. Rashidirad et al., 2013;
Zhang et al., 2012; Jayaram et al., 2010), management accounting (e.g. Al-Omiri and
72
Drury, 2007; Auzair and Langfield-Smith, 2005) and strategic management (e.g. Ward et
al., 2007; Lei et al., 1996). CT adopts the premise that any organisational, management
and operations system cannot be equally applicable and/or effective in all contexts and
environments (Drazin and Van de Ven, 1985). Therefore, a specific context can be more
conducive for a specific system than other contexts, which positions the concept of fit at
the heart of CT (Drazin and Van de Ven, 1985). However, how the fit is defined by
researchers can have a profound effect on the type of theory developed, data collection
and the statistical analysis needed for testing the theory (Al-Omiri and Drury, 2007;
Drazin and Van de Ven, 1985). To clarify the confusion surrounding the
operationalisation of fit within the CT, different approaches have been suggested to define
“fit” (Venkatraman, 1989; Drazin and Van de Ven, 1985). Drazin and Van de Ven (1985)
have introduced three approaches for fit, namely the selection approach, the interaction
approach and the system approach while Venkatraman (1989) has highlighted the
moderation and mediation perspectives within the CT. Each of these different approaches
is explained in detail below.
3.3.2.1 The selection approach
The selection approach perceives fit as a relationship or congruence between
organisational context (e.g. technology, size, or environment) and organisational structure
(e.g. formalisation, complexity) without paying any attention to the potential impact of
that congruence on firm performance (Drazin and Van de Ven, 1985). This lack of
attention to the impact of context-structure fit on performance has been justified using the
natural and managerial perspectives (Drazin and Van de Ven, 1985).
The natural perspective contends that only high-performing organisations survive due to
their continuous and gradual adaptation or fit to environment (Drazin and Van de Ven,
1985). Given this notion, it is considered sufficient to test only the association between
context and structure without the need to include performance explicitly. The managerial
perspective goes beyond the natural perspective by taking into consideration organisation
design at both macro- and micro-level (Drazin and Van de Ven, 1985). This perspective
assumes constraints are imposed by macro levels of an organisation on its micro levels
73
preventing the latter from fully adopting the structural design most appropriate for their
particular conditions. Consequently, all structural variables constrained by macro levels
can be analysed for their fit with context using the selection approach, while the structural
variables not constrained can interact with context to predict variation in performance and
so need to be analysed through the interaction approach (Drazin and Van de Ven, 1985).
3.3.2.2 The interaction approach
In contrast to the selection approach explained above, the interaction approach perceives
fit as the interaction between two variables (e.g. context and structure) to explain variation
in a third variable (e.g. performance) (Drazin and Van de Ven, 1985). Consequently, in
this approach understanding the congruence between context and structure is not of high
importance as it was the case in the selection approach. It is mainly the variation in
performance that can be explained by the interaction between context and structure that is
of importance in this approach (Drazin and Van de Ven, 1985). The interaction approach
as a way of operationalising fit is widely spread in the academic literature although mixed
results have been obtained in relation with the multiplicative term formed to represent the
interaction in the regression equation (Luft and Shields, 2003; Drazin and Van de Ven,
1985). A considerable number of researchers have adopted this perspective in the
reviewed literature (e.g. Dabhilkar and Åhlström, 2013; Birdi et al., 2008; Das and
Jayaram, 2007; Patterson et al., 2004; Challis et al., 2002). This perspective will also be
used in this research to test for the proposed interaction between lean service practices.
3.3.2.3 The system approach
The system approach criticises the previous two approaches to fit on the ground that they
adopt a reductionism approach assuming that an organisation can be decomposed into
several elements that can be investigated independently (Drazin and Van de Ven, 1985).
More specifically, the selection and interaction perspective of fit tend to examine fit
between single contextual variables and single structural variables and how each pair of
variables interact to predict variation in performance (Drazin and Van de Ven, 1985). The
results of each examination are then aggregated to make conclusions about the whole
organisational system (Drazin and Van de Ven, 1985). To avoid the reductionism
74
problem, the system approach advocates a multivariate analysis in which the fit between
several contingent, structural and performance variables is addressed simultaneously
(Miller, 1981).
3.3.2.4 The moderation perspective
The moderation perspective within CT implies that a relationship between one
independent variable and one dependent variable is dependent on the level of a third
variable called “moderator” (Frazier et al., 2004; Venkatraman, 1989). Figure 3-1 depicts
this type of relationship. However, a moderator can either moderate the form or strength
of the assumed relationship where understanding the type of moderation is critical to
determine the appropriate statistical analysis needed to detect it (Frazier et al., 2004;
Venkatraman, 1989). Depending on the theoretical argument, a particular moderation form
can be relied on along with the appropriate statistical analysis to test it.
Figure 3-1: The moderation perspective
3.3.2.5 The mediation perspective
The mediation perspective represents a case where the relationship between a predictor
and a criterion variable can be either completely or partially explained by a third variable
called “mediator” (Frazier et al., 2004). This particular case is represented in Figure 3-2
where the variable C is a mediator variable. A mediator C can either account fully for the
A-B association resulting in a complete mediation, or only decrease the A-B association
resulting in a partial mediating effect.
Independent
variable
Dependent
variable
Moderator
75
The moderation and mediation perspectives represent two different theoretical concepts as
clear from the explanation above. Therefore, a specific variable can be either a moderator
or a mediator depending on the conceptual argument of the research, but the same variable
can be conceptualised as a moderator in one study and as a mediator in another (Frazier et
al., 2004).
Figure 3-2: The mediation perspective
3.4 Theoretical model
The theoretical model developed in this present research makes use of the two well known
theories discussed above, namely STS and CT. More specifically, lean service is viewed
in this research as a socio-technical system following the perspective of Shah and Ward
(2007, 2003). Therefore, lean service practices are classified into two categories or sides,
technical side and social side as presented in Tables 3-1 and 3-2. By viewing lean service
as a socio-technical system, the mechanism of the STS and the existing lean literature can
then be used to develop an argument through which the main effect of each side and the
expected synergy between the two sides in influencing firm performance are highlighted.
In line with the STS perspective, the theoretical model proposed in this study assumes a
direct positive impact of each side of lean service on both operational and financial
performance of adopters. Moreover, this model expands the traditional lean-performance
model by including the likely synergistic (i.e. non-additive) effect resulting from the
Independent
variable (A)
Dependent
variable (B)
Mediator
(C)
76
ability of each side of lean service to enhance the influence of the other side on firm
performance. By doing so, the model in this research attempts to overcome limitations of
previous studies that failed to uncover the full potential of lean service given their neglect
to the non-additive impact of each side on performance (see section 2.4).
On the other hand, the CT, specifically the mediation perspective, will be essentially used
to support the theoretical argument in relation to the effect of two contextual variables
(ABC and business strategy) on LTPs. More specifically, by adopting the mediation
perspective, the model suggests a direct positive impact of ABC on LTPs which, in turn,
have an effect on performance. This highlights the possible indirect influence of ABC on
performance through its effect on other organisational capabilities (LTPs) which are
expected to affect performance. In addition, business strategy is expected to have a dual
effect on LTPs directly and indirectly through its direct impact on ABC which highlights
the core role of the accounting system in the case of lean service.
Finally, to provide robust findings in connection with the impact of lean service on firm
performance, 3 contextual variables are included in the suggested model as control
variables given their expected direct effect on firm performance. These variables include
firm size, firm age and ABC. Figure 3-4 reflects this argument and depicts the theoretical
model that will be tested in this present research study.
3.5 Research hypotheses
Based on the theoretical model presented in Figure 3-3, a number of hypotheses are
developed and then empirically tested. These hypotheses can be divided into two groups.
The first group of hypotheses (H1a-H6b) relates to the impact of lean service and
contextual variables on firm operational and financial performance. The second group of
hypotheses (H7-H11) relates to the effect of ABC and business strategy on the technical
side of lean service. The research hypotheses in the order just mentioned are developed
below.
77
Figure 3-3: The theoretical model3
3.5.1 The impact of lean service and contextual variables on firm performance
3.5.1.1 Lean technical practices (LTPs) and firm performance
LTPs presented in Table 3-1 are introduced to improve customer value by identifying,
measuring and eliminating non-value adding activities (NVAs) from processes (Ehrlich,
2006; Womack and Jones, 1996). By doing so, several benefits can be expected from
those practices to adopters. The lean service literature, consisting mainly of conceptual
and case studies, highlights such benefits as ameliorating lead and cycle time, increasing
efficiency, improving flexibility, improving customer satisfaction and profitability (see
Table 3-3 for a comprehensive list) (Staats et al., 2011; Piercy and Rich, 2009b; Bhatia
and Drew, 2007; Abdi et al., 2006; Swank, 2003; Yavas and Yasin, 2001). However, the
3 This theoretical framework was presented to and accepted by the International Journal of Operations and
Production Management.
ABC LTPs
Cost leadership
strategy
Operational
Performance
Financial
Performance
LSPs
Size (+/-);
Age (+/-);
ABC (+)
Differentiation
strategy
LTPs * LSPs
H9 (+)
H10 (-)
H8 (+)
H2a (+)
H2b (+)
H1a (+)
H1b (+)
H3a (+)
H3b (+)
H11 (+)
H4a (+/-)
H5a (+)
H6a (+/-)
H4b (+/-)
H5b (+)
H6b (+/-)
H7 (+)
78
lack of systematic application of LTPs is argued to result in no improvement (Kim et al.,
2012; Allway and Corbett, 2002), and in effect, such systematic application is not easy to
attain (Robinson and Schroeder, 2009; Scherrer-Rathje et al., 2009; Bhasin, 2008; Maskell
and Kennedy, 2007; Atkinson, 2004). Therefore, it is not inevitable that all adopters of
LTPs can be expected to achieve the purported benefits of those practices (Bhasin, 2008).
This inconclusive conclusion is supported by the contrasting results reported by empirical
studies, mainly in manufacturing, which focus on the effectiveness of the LTPs. Some
empirical evidence proves the theoretical argument of a positive relation between LTPs
and performance (e.g. Dabhilkar and Åhlström, 2013; Shah and Ward, 2003; Cua et al.,
2001). In contrast, no relation between LTPs and performance is also documented (e.g.
Birdi et al., 2008; Patterson et al., 2004). However, as a larger number of researchers have
proposed and supported a direct positive relation between LTPs and firm performance
indicators shown in Table 3-3, the following hypotheses are reported:
H1a: There is a direct positive relationship between LTPs and operational performance of
service firms.
H1b: There is a direct positive relationship between LTPs and financial performance of
service firms.
Examining the impact of LTPs on operational and financial performance separately is
important. In manufacturing, even a successful lean attempt may result in deterioration in
net profit corresponding to liquidating high levels of inventory stored prior to
implementing lean manufacturing (Meade et al., 2010). Liquidating inventory transfers the
capitalised value of inventory to expenses charged to the year in which lean system
requires reducing the amount of inventory acquired. Therefore, although lean system is
successfully attacking one type of waste (inventory) to improve processes, it may lead to a
reduction in net profit until the level of inventory has stabilised, and then an increase in
profit can be expected (Meade et al., 2010). However, in most service industries there is
no or low levels of inventory (Apte and Goh, 2004; Lowry, 1990; Dearden, 1978). But the
argument here is about whether operational improvements obtained from lean service can
overcome any costs (e.g. training sessions, CFL) associated with the adoption of lean
79
service. Consequently, having no materialised financial benefits should not be the only
indication of lean failure. Operational benefits should be considered as well before
doubting the successfulness of lean service.
3.5.1.2 Lean social practices (LSPs) and firm performance
Given the view of lean service as having two sides (i.e. LTPs and LSPs), the impact on
performance does not seem to be restricted to LTPs only. LSPs are found to be
independently capable of explaining variation in the performance of adopters (De
Menezes et al., 2010; González-Benito, 2005; Patterson et al., 2004; Shah and Ward,
2003; Cua et al., 2001). For instance, investing in training programs is believed to advance
the quality of employees by improving their current skills and helping acquire new skills
so that they become multi-functional employees able to perform various tasks and serve in
different locations when needed (Delaney and Huselid, 1996; Huselid, 1995).
Consequently, a multi-skilled employee can smooth operations processes when a
bottleneck appears at any point of a process by helping other employees working in that
part of the process. However, multi-skilled employees like others need to be motivated
and empowered to utilise their skills in this way (Liker and Morgan, 2006; Delaney and
Huselid, 1996; Huselid, 1995). Therefore, an appropriate rewarding system that can align
the interest of employees with that of an organisation along with decentralisation in the
decision making process will be effective in motivating employees to achieve pre-
specified goals (Delaney and Huselid, 1996; Huselid, 1995). In contrast to this positive
view, there is little evidence indicating a lack of ability of LSPs to generate significant
benefits to adopters (Bonavia and Marin, 2006). These mixing results are not confined to
the operations-oriented literature; empirical evidence from the human resource literature
also provides different views. While early evidence has proved a positive relation between
HRMs and firm operational and financial performance (Delaney and Huselid, 1996;
Huselid, 1995), later evidence has failed to capture such positive relation and has
concluded that job rotation has negatively affected productivity and collectively HRMs
have increased labour cost (Cappelli and Neumark, 2001). However, given that greater
evidence suggests a direct positive relation between LSPs and performance, the following
hypotheses are reported:
80
H2a: There is a direct positive relationship between LSPs and operational performance
of service firms.
H2b: There is a direct positive relationship between LSPs and financial performance of
service firms.
These hypotheses are highly important in the case of lean service for managements who
get excited about the possible benefits that can be achieved from LTPs so that they rush to
implement LTPs without realising the likely role of LSPs in independently improving
performance.
3.5.1.3 The synergistic effect of LTPs and LSPs on performance
In addition to the direct relation proposed between LTPs and LSPs on the one hand, and
firm performance on the other hand, the mechanism of the STS presented in Figure 3.4
suggests another indirect (non-additive) role of both LTPs and LSPs in improving firm
performance (Dabhilkar and Åhlström, 2013; Kull et al., 2013; Das and Jayaram, 2007;
Trist, 1981). The STS indicates that the best outcome of any socio-technical system can
only be achieved by simultaneous emphasis on implementing practices from both
subsystems (i.e. LTPs and LSPs) (Dabhilkar and Åhlström, 2013; Trist, 1981). This
implies that there can be a synergy between LTPs and LSPs where the effectiveness of
each set of practices on performance is enhanced by the other. For instance, in the case of
lean service, the value stream mapping (VSM) can be applied to identify NVAs and
bottlenecks. Inarguably, some of these deficiencies could be eliminated by untrained
employees. It could be argued that trained and multi-skilled employees would be more
effective in elimination of NVAs and bottlenecks if empowered to do so. However, if the
NVAs were not identified, trained and multi-skilled employees would carry out all
activities efficiently but not necessarily effectively. The gain achieved in each of the
forgoing cases will be less than the gain that could be obtained in a third case where VSM
is initially implemented by a company who has trained multi-skilled employees to
improve its processes. Adopting this synergy perspective implies that each set moderates
the form of relationship between the other set and performance, and therefore the
traditional moderation perspective cannot be adopted here as it is not possible to determine
81
which set represents the independent variable and which one represents the moderator
variable (Sharma et al., 1981). Some researchers have empirically examined the synergy
perspective with high focus on the synergy between combinations of JIT, TQM and
HRMs (Furlan et al., 2011; Cappelli and Neumark, 2001; Flynn et al., 1995). A few others
who have focused on the wider definition of lean as provided by Shah and Ward (2003)
used a limited number of practices to represent each bundle and reported inconclusive
results on the synergistic effect (Dabhilkar and Åhlström, 2013; Birdi et al., 2008; Das and
Jayaram, 2007; Patterson et al., 2004). While some studies forwarded positive indications
on the presence of synergy (Birdi et al., 2008; Das and Jayaram, 2007), others failed to
verify the assumed synergy (Dabhilkar and Åhlström, 2013; Patterson et al., 2004).
However, the hypotheses are formulated in favour of the theoretical argument proposed by
the STS as follows:
H3a: There is a synergy between LTPs and LSPs in improving operational performance
of service firms.
H3b: There is a synergy between LTPs and LSPs in improving financial performance of
service firms.
Figure 3-4: The mechanism of the socio-technical system theory Adapted from: Manz and Stewart (1997)
Technical side of
lean service:
VSM, root cause
analysis, 5Ss, etc.
Social side of lean
service:
Training, employee
involvement and
empowerment, etc.
Organisational
performance:
Quality, productivity,
flexibility,
profitability etc.
82
3.5.2 The effect of contextual variables on performance and/or lean service
There can be several factors that may affect either companies’ performance (Capon et al.,
1990) or lean adoption or both (Shah and Ward, 2003). Several researchers in the
literature of lean manufacturing have stressed the importance of contextual variables in
determining the lean-performance association such as the nature of process, firm size and
firm age (Shahrukh, 2011; Malladi et al., 2010; Pont et al., 2008; Shah and Ward, 2003;
Cua et al., 2001; Christopher, 2000). Not fully taking into account the effect of such
contextual variables might have been behind inconsistent empirical results concerning the
lean- performance relationship (Staats et al., 2011; Shah and Ward, 2003). To identify the
key potential contextual variables, the lean service literature, lean manufacturing
literature, and the research focusing on the performance of the service sector including
management accounting literature and business strategy literature were searched.
Surveying the aforementioned different literatures, four important contextual variables
have been identified and presented below.
3.5.2.1 Firm age and firm performance
Firm age can also play a role in determining the overall firm performance. Older firms can
be assumed to have higher experience in running business in comparison with young firms
(Coad et al., 2013; Glancey, 1998). Consequently, their accumulated knowledge and
experience put them in a better position to run their operations more efficiently than less
experienced firms leading to better performance (Coad et al., 2013; Lundvall and Battese,
2000; Glancey, 1998). In addition, older firms can rely on their established reputation, if
exists, to achieve higher margins than those earned by new or young firms whose
reputation has not been well established (Glancey, 1998). However, older firms may suffer
from rigidity and inflexibility in responding to market changes and adopting new
innovations which may negatively impact their productivity and responsiveness (Coad et
al., 2013; Wagner et al., 2012; González-Benito, 2005; Shah and Ward, 2003). These two
contrasting theoretical arguments in connection with the impact of age on performance
have been supported by the mixing results reported from empirical studies.
83
Yasuda (2005) finds a negative relationship between firm age and firm growth based on a
large sample of Japanese manufacturing firms. Lundvall and Battese (2000) report
evidence based on data from 235 Kenyan manufacturing firms on the impact of firm size,
age and technical efficiency. Their analysis detects a positive effect of firm size in two
industries out of three, but similar effect has not been proved for firm age except for one
industry (i.e. textile). Coad et al. (2013) provide empirical evidence from a panel of
Spanish manufacturing firms suggesting a positive relation between firm age and
performance but with diminishing rate. González-Benito (2005) shows by surveying 186
manufacturing companies that there is no relationship between operational and financial
performance (represented by cost, quality, flexibility, reliability and speed and return on
assets, respectively) and company size, type of industry and age of equipment.
Furthermore, the results of Shah and Ward (2003) convey a significant negative effect of
firm size and age on operational performance. Wagner et al. (2012), using empirical data
from 259 manufacturing firms, prove a positive influence of firm size on financial
performance while firm age is found to have no influence. Consequently, given this
contradiction in theory and empirical findings on the nature of age-performance
association, the following non-directional hypotheses are formulated:
H4a: There is a direct relationship between firm age and operational performance of
service firms.
H4b: There is a direct relationship between firm age and financial performance of service
firms.
3.5.2.2 Activity-based costing system and firm performance
The extensive competition facing service firms reduces the power of service firms to set
arbitrary prices for their services as they used to do few years ago (Yu-Lee, 2011;
Karmarkar, 2004). This underpins the need to focus more on cost and the importance of
cost related information (Guilding et al., 2005). The nature of services further signifies
this importance. Broadly speaking, services cannot be inventoried for later use and a
significant amount of cost in service firms is fixed at least in the short term (Carenys, and
Sales, 2008; Dearden, 1978). Therefore, if a sale transaction is not made, the associated
84
revenue is lost forever and the time-based overhead cost of the period will be added to that
loss (Schlissel and Chasin, 1991). Consequently, having an advanced costing system to
provide accurate information on service cost is essential for management to measure
resources used, evaluate operating performance, and make informed decisions and price
its products/services (Martinson, 2002; Clarke and Mullins, 2001; Kock, 1995; Hegde and
Nagarajan, 1992; Anania, 1987).
Using data from 280 manufacturing and service firms, Guilding et al. (2005) demonstrate
the increasing importance of cost information for pricing products/services for companies
facing extensive competition and/or operating in the service sector. In their empirical
study of 61 business unit managers, Mia and Clarke (1999) prove the positive and direct
role of information provided by the MAS in improving firm operational and financial
performance. Chong and Chong (1997) also verify the direct positive influence of the
broad scope MAS information on firm performance using data from 62 Australian
business managers. In a similar vein, Gerdin et al. (2005) suggests, based on data from
132 production managers, that the use of greater amounts of MAS information is
positively related to firm performance. However, the results show no support to the
assumed positive effect of the more frequent use of such information on firm performance.
Pizzini (2006) provides more supportive evidence by studying the relationship between
cost- design systems, managers’ beliefs about the relevance and usefulness of cost data
and financial performance using a sample of 277 US hospitals. The results indicate that
managers consider cost data to be relevant and useful if it is provided with greater detail,
classified according to behaviour and reported more frequently. In addition, reporting cost
data with greater detail is proved to be associated with financial indicators namely,
operating margin, cash flow and administrative expense.
Among the MASs currently available, traditional costing systems (TAS) including
absorption costing provide a narrow range of cost information by relying on standard
costing and assigning overhead costs based on volume-based factors (Sheu and Pan, 2009;
Baggaley, 2006; Toomey, 1994; Chenhall and Langfield-Smith, 1998). In contrast to the
traditional methods, activity-based costing (ABC) allows for the provision of more
85
detailed, useful and accurate cost information by linking costs to activities that cause those
costs to occur (Khataie and Bulgak, 2013; Mishra and Vaysman, 2001; Adams, 1996).
Mishra and Vaysman (2001) examine the capability of ABC and TAS in improving the
profitability of adopters. They find that ABC can lead to higher profitability when the
uncertainty surrounding the managers’ private information is high. However, when the
level of uncertainty is low, TAS is capable of generating higher profits for adopters. For
this reason Mishra and Vaysman (2001) conclude that ABC and TAS will co-exist in an
economy. Similarly, Sheu and Pan (2009) investigate the level of profitability expected
from ABC and TAS under high and low levels of knowledge uncertainty in a virtual
enterprise of R&D activities. Like Mishra and Vaysman (2001), Sheu and Pan (2009)
convey that ABC generates higher level of profits when the knowledge uncertainty
surrounding the R&D member is high while TAS becomes superior in cases where this
uncertainty is low.
Shields (1995) surveys 143 firms to investigate the level of success of ABC and the
implementation variables associated with the level of success. The results indicate that
adopters of ABC obtain a moderate level of success and attain financial benefits from
adopting ABC. The success and financial benefits from ABC are found to be associated
with six behavioural and organisational variables, namely top management support,
linkage to competitive strategy especially quality and JIT/speed, linkage to performance
evaluation and compensation, training in implementing ABC, non-accounting ownership,
and adequate resources. Cagwin and Bouwman (2002) concentrate on the financial
performance outcome of ABC and the conditions under which this outcome is improved.
Employing data from 106 manufacturing and 98 non-manufacturing firms, Cagwin and
Bouwman (2002) report empirical evidence suggesting no direct effect of ABC on ROI.
However, a positive relationship between ABC and improvement in ROI is detected when
ABC is adopted in complex and diverse firms, when cost information is important, when
ABC is implemented concurrently with other initiatives like JIT and TQM, and when
ABC is implemented by manufacturing firms. Kennedy and Affleck-Graves (2001)
examine whether a sample of UK firms adopting ABC outperform their counterpart in
86
terms of financial performance. Their findings reveal that ABC firms achieve around 27%
higher abnormal return than non-ABC firms. Finally, Ittner et al. (2002) also report
empirical evidence on the direct positive impact of ABC on quality levels and cycle time.
As a result of the above illustration, it is expected in this study that better operational and
financial performance is related to the use of ABC.
H5a: There is a direct positive relationship between the use of ABC and operational
performance of service firms.
H5b: There is a direct positive relationship between the use of ABC and financial
performance of service firms.
3.5.2.3 Firm size and firm performance
Firm size has usually been considered an important determinant of firm performance.
Large firms are argued to have more financial and human resources and therefore they
may enjoy the benefits of economies of scale leading to better performance (Jayaram et
al., 2010; Shah and Ward, 2003; Glancey, 1998). In addition, the higher level of resources
available for large firms can be advantageous by allowing for more experimentation with
new technologies and innovations that may improve their productivity and efficiency
(Coad et al., 2013; Wagner et al., 2012; Galende and de la Fuente, 2003; Shah and Ward,
2003).
In contrast, large firms can be more reluctant or slower in adopting innovative methods
and techniques that are capable of improving their performance given the complexity of
their operations (Shah and Ward, 2003; Hannan and Freeman, 1984). This reluctance
highlights the possible negative effect of firm size on performance. These two contrasting
theoretical arguments in connection with the impact of size on performance have been
supported by the mixing results reported from empirical studies.
Yasuda (2005) finds a negative relationship between firm size and firm growth based on a
large sample of Japanese manufacturing firms. Lundvall and Battese (2000) report
evidence based on data from 235 Kenyan manufacturing firms on the impact of firm size,
87
age and technical efficiency. Their analysis detects a positive effect of firm size in two
industries out of three, but similar effect has not been proved for firm age except for one
industry (i.e. textile). González-Benito (2005) shows by surveying 186 manufacturing
companies that there is no relationship between operational and financial performance
(represented by cost, quality, flexibility, reliability and speed and return on assets,
respectively) and company size, type of industry and age of equipment. Furthermore, the
results of Shah and Ward (2003) convey a significant negative effect of firm size and age
on operational performance. Wagner et al. (2012), using empirical data from 259
manufacturing firms, prove a positive influence of firm size on financial performance
while firm age is found to have no influence. Consequently, given this contradiction in
theory and empirical findings on the nature of size-performance association, the following
non-directional hypotheses are formulated.
H6a: There is a direct relationship between firm size and operational performance of
service firms.
H6b: There is a direct relationship between firm size and financial performance of service
firms.
3.5.2.4 Business strategy and lean technical practices
The core of lean service practices is to improve the value delivered to customers and
respond more quickly to their increasingly changing needs (Shah and Ward, 2003;
Womack and Jones, 1996). With this objective, lean service goes in line with the
objectives of a differentiation strategy (Kennedy and Widener, 2008). Differentiators
operate in an uncertain environment where customer needs change constantly and they are
likely to experience lower levels of productivity because of their higher levels of
products/services customisation (Kumar and Telang, 2011; Gosselin, 1997; Bruggeman
and Stede, 1993). It follows that differentiators can be motivated to implement lean
practices that have proved effective in developing new products/services quicker and with
high quality to respond faster to changes in customer needs (Qudrat-Ullah et al., 2012;
Cooper and Maskell, 2008; Kennedy and Widener, 2008; Karlsson and Ahlstrom, 1996).
In addition, although differentiators are usually able to ask for price premium to
88
compensate for any extra cost incurred, they do not completely ignore their operating
costs and this can be seen from their tendency to implement ABC for more accurate cost
information (Chenhall and Langfield-Smith, 1998). This can be an additional impetus for
differentiators to use lean practices in order to exert some control over their operating
costs by eliminating non-value added activities from their processes (Qi et al., 2011).
On the other hand, the mechanism through which lean service is assumed to improve
customer’s value is the identification and elimination of non-value adding activities
(NVA) which usually leads to cost reduction and efficiency improvement (Shah and
Ward, 2007, 2003; Womack and Jones, 1996). Because of that lean service has been
perceived by many companies as a cost reduction program although cost reduction is an
outcome rather than a main objective (Hartwell and Roth, 2010; Atkinson, 2004). In
consequence, cost leaders, who usually compete on the price base and therefore they seek
cost reduction and efficiency in all possible areas (Ward et al., 2007; Frey and Gordon,
1999, Porter, 1980), are anticipated to also have a strong motivation to implement this
system (Qi et al., 2011).
Only a few researchers have endeavoured to empirically examine the aforementioned
argument and reported mixing results. Among those few researchers, Chenhall and
Langfield-Smith (1998), using empirical data from 78 manufacturing firms, highlights the
positive fit between lean practices and differentiation strategy where better performance is
achieved when lean practices are used under the differentiation strategy. In addition,
Baines and Langfield-Smith (2003) use data from 140 manufacturing firms and forward
evidence suggesting that adopting differentiation strategy will lead to increased use of
Just-in Time (JIT) and Total Quality Management practices (TQM) considered part of the
lean system (Shah and Ward, 2003). In a similar vein, Ward et al. (2007) report empirical
evidence revealing significant differences on some operations decisions between three
distinctive business strategies, namely price leaders, differentiators and broad-based
competitors that combine the focus of the first two strategies. The results indicate that
broad-based competitors are characterised with the highest implementation level of JIT
and quality practices while no clear difference between differentiation strategy and price
89
leaders could be detected. In contrast to the above literature, Qi et al. (2011) demonstrate
through an empirical study of 604 Chinese manufacturing companies that companies
adopting cost leadership strategy are more likely to implement lean system compared to
differentiators. In summary, the empirical evidence from the few studies found in the
literature suggests a direct positive relation between differentiation strategy and lean
practices while the empirical evidence is rather weak for the cost leadership strategy.
However, as mentioned before, cost leaders can, at least theoretically, be motivated to
make use of and benefit from lean practices too. Therefore, the following hypotheses from
a theoretical perspective are reported:
H7: There is a direct positive relation between differentiation strategy and the
implementation of LTPs.
H8: There is a direct positive relation between cost leadership strategy and the
implementation of LTPs.
3.5.2.5 Business strategy and activity-based costing system
Given the dissimilarities between the two types of strategies (differentiators/prospectors
vs. cost leaders/defenders) highlighted before, the accounting system used by
prospectors/differentiators on the one hand and defenders/cost leaders on the other hand is
expected to differ (Porter, 1980; Miles and Snow, 1978). Hammad et al. (2010) propose
that business strategy adopted will have a direct effect on the design and functionality of
MAS. Gosselin (1997) points out that as prospectors/differentiators operate in highly
uncertain environment compared to defenders/cost leaders; they usually require a broader
scope of information which can be fulfilled through the use of ABC. Moreover,
prospectors usually devote more resources to experiment with innovations and
consequently they are more likely to invest in innovative systems such as ABC than
defenders (Chenhall, 2003; Chenhall and Langfield-Smith, 1998; Gosselin, 1997).
Further, as differentiators compete by producing more customised products/services to be
perceived as unique by customers (Ward et al., 2007; Bruggeman and Stede, 1993; Porter,
90
1980), these companies may prefer ABC given its superiority over TAS in generating
unbiased cost information on their various and more customised products/services
(Khataie et al., 2011; Cagwin and Bouwman, 2002; Gosselin, 1997; Cooper and Kaplan,
1992). Conversely, as TAS can provide less distorted cost information for companies with
a relatively limited group of standardised products/services and at lower cost, cost leaders
may be more likely to rely on TAS rather than ABC (Lamminmaki and Drury, 2001;
Gosselin, 1997). Although Abdel-Kader and Luther (2008) find no difference between
cost leadership and differentiation strategies in terms of the level of their MAS
sophistication, several researchers have documented a strong effect of the type of strategy
adopted on the MAS in use as presented below.
Baines and Langfield-Smith (2003) use data from 140 manufacturing firms and forward
evidence suggesting that adopting differentiation strategy will lead to increased use of
advanced MASs including ABC, target costing and life cycle costing. Amir et al. (2010)
examine the impact of service process type, business strategy, intensity of competition on
the use of advanced performance measurement system (PMS). They find that firms
adopting differentiation strategy are more likely to use advanced PMS than firms adopting
low cost strategy regardless of being mass or professional services. Chenhall (2003) finds
that differentiators require an organic MCS with broad scope while cost leaders require a
mechanistic MCS. In addition, he proposes that organic organisational structures perceive
ABC to be highly effective in contrast to the mechanistic structure. Based on data
collected from 121 UK service firms, Auzair and Langfield-Smith (2005) support the
findings of Chenhall (2003) by concluding that mass services and cost leaders are more
likely to adopt more formal MCS –which does not support ABC- while professional
services and differentiators will adopt informal MCS which supports ABC. Gosselin
(1997) uses data from 161 manufacturing firms and examines the effect of business
strategy and organisational structure on the adoption of activity management (AM)
decomposed into activity analysis (AA), activity cost analysis (ACA) and ABC. The
author demonstrates that prospectors are more likely to adopt ABC than defenders. Frey
and Gordon (1999) reveal through empirical evidence that the choice of business strategy
does not influence the decision to adopt ABC. However, they find that differentiators
91
achieve higher level of benefits from ABC than cost leaders. Similarly, Chenhall and
Langfield-Smith (1998) prove empirically that ABC benefits differentiators more than
cost leaders because they use ABC not only to manage costs but also to understand value-
added activities that enhance product differentiation.
H9: The differentiation strategy is positively related to the use of ABC
H10: The cost leadership strategy is negatively related to the use of ABC.
3.5.2.6 Activity-based costing system and lean technical practices
The importance of ABC in the lean context stems from its ability to overcome the
limitations of TAS which have been argued to affect the implementation of lean service
practices (Banker et al., 2008; Datar et al., 1991). Several conceptual and case studies
have discussed the role of the accounting system in the lean context (e.g. Chiarini, 2012;
Cooper and Maskell, 2008; Kennedy and Widener, 2008; Grasso, 2005; Karlsson and
Ahlstrom, 1996; South, 1993; Datar et al., 1991). In this body of literature, there is almost
a consensus among researchers on the detrimental effect of TAS on the implementation of
lean practices. The core argument is that lean service seeks to identify and eliminate non-
value adding activities in order to improve processes (Shah and Ward, 2007; 2003;
Womack and Jones, 1996). However, the main focus of TAS is on the allocation of all
overhead costs to cost objects, and therefore it generates aggregate cost information. This
leads to waste being hidden in the overhead allocation rate and conceals areas which
require improvements (Maskell, 2006; Chenhall and Langfield-Smith, 1998; Toomey,
1994).
According to Datar et al. (1991), TAS may stifle process improvement initiatives by
providing inaccurate cost information to decision makers which sends incorrect signals in
relation to the effectiveness of a process innovation leading decision makers to mistakenly
cease such innovation. For example, this can happen when TAS, due to its distorted
information, reveals that a product/service A is profitable when in fact it is not while a
product/service B is not profitable when in fact it is. In such scenario, a company may not
92
attempt to improve the process of product/service A because it is seen to be profitable
while the company may decide to apply lean practices to improve the process of
product/service B. In this case, the need to improve the process of product/service A is
concealed. In addition, the attempt to improve product/service B may not result in the
desired improvement leading to questioning the effectiveness of the improvement program
implemented. Similar examples have been reported by several researchers who have
explicated how some companies have been about to cease the implementation of lean
practices because of the misleading information generated from their TAS (e.g. Cooper
and Maskell, 2008; Datar et al., 1991; Turney and Anderson, 1989).
As lean system focuses on processes which consist of a series of activities, it is argued that
lean system can be supported by an activity-oriented costing system (Khataie and Bulgak,
2013). ABC breaks down a process into its constituent activities and traces overhead costs
to these activities, and then allocates the accumulated overhead costs to cost objects
through both volume-based and non-volume based cost drivers (Askarany et al., 2010;
Banker et al., 2008; Cooper and Kaplan, 1992). Based on that, ABC is believed to
improve the visibility of what truly drives cost and consumes resources which results in
more accurate cost information (Ruiz-de-Arbulo-Lopez et al., 2013; Schoute, 2011;
Spedding and Sun, 1999; Cooper and Kaplan, 1992). As a result, ABC is thought to
overcome the limitations of TAS and support lean service in different ways.
First, because ABC accumulates costs at activity level, it helps differentiate between value
adding and non-value adding activities, which is critical for lean service (Banker et al.,
2008; Maiga and Jacobs, 2008; Larson and Kerr, 2007; Clarke and Mullins, 2001;
Gunasekaran and Sarhadi, 1998). In addition, highlighting the cost of non-value adding
activities can be a critical factor for justifying the need to adopt lean practices in order to
eliminate such cost (Ittner et al., 2002). Ittner (1999) explicates how ABC can be used to
prioritise quality improvement efforts. Second, cost information from ABC is argued to
result in more realistic budgets and variance analyses exposing the true negative variances
(Abu Mansor et al., 2012; Grasso, 2005; Innes and Mitchell, 1995; Cooper and Kaplan,
1992). Consequently, implementing lean practices to improve those negative variances is
93
less likely to yield a disappointing outcome which usually leads to questioning the
effectiveness of lean practices and finally abandoning them (Banker et al., 2008; Larson
and Kerr, 2007; Datar et al., 1991).
In favor of the above argument, Innes and Mitchell (1995) survey the largest UK
companies and find them to depend on the measures generated from ABC to support other
improvement initiatives such as continuous improvement, TQM and JIT. Adam (1996)
argues that when ABC is used in conjunction with other process improvement systems
such as quality and lean system, companies enjoy a higher level of benefits. Along this
line, Khataie and Bulgak (2013) use system dynamics modeling tool and reveal the
importance of ABC in achieving the aim of lean system. Another simulation study by Li et
al. (2012) also shows that ABC has superiority over TAS in bridging the gap between
operational and financial improvements of lean companies. Adopting a case study
methodology, Chiarini (2012) demonstrates the advantages of ABC in a medium-sized
lean firm. Finally, Banker et al. (2008) report empirical evidence indicating that
manufacturing companies adopting ABC are more likely to adopt lean practices in their
operations. Therefore, the following hypothesis is formulated:
H11: There is a positive relation between the use of ABC and the implementation of LTPs.
3.6 Summary of chapter 3
In this chapter all constructs required to develop the theoretical model attempting to
overcome limitations in the existing literature have been identified. The two core theories
(i.e. STS and CT) guiding and supporting the developed model of this study have been
discussed and the way and importance of their use in the current study have been
explained and justified. The theoretical model views lean service as a socio-technical
system with two distinctive but interdependent sides. Consequently, the 54 lean service
practices identified through the systematic review of lean service literature have been
classified into those that represent the technical side of the system (i.e. 37 practices as in
Table 3-1) and those symbolising the social side of the system (i.e. 17 practices as in
Table 3-2). Based on the STS and the lean service literature, the model expects a direct
94
positive influence of each side on firm performance. In addition, the two sides are
expected to synergistically interact to improve firm performance over and above the level
resulted from each side separately. Furthermore, four contextual variables have been
identified due to their expected direct and/or indirect influence on LTPs and/or
performance. These variables include ABC, business strategy and firm size, firm age.
Finally, the model of this study has generated a number of hypotheses (summarised in
Table 3-4) which link together the various constructs of the model and will be the input
for chapter 5 which focuses on the empirical verification of those hypotheses.
95
Table 3-4: Summary of the research hypotheses
No. Hypotheses Expected
result
H1a There is a direct positive relationship between LTPs and
operational performance of service firms. +
H1b There is a direct positive relationship between LTPs and financial
performance of service firms. +
H2a There is a direct positive relationship between LSPs and
operational performance of service firms. +
H2b There is a direct positive relationship between LSPs and financial
performance of service firms. +
H3a There is a synergy between LTPs and LSPs in improving
operational performance of service firms. +
H3b There is a synergy between LTPs and LSPs in improving financial
performance of service firms. +
H4a There is a direct relationship between firm age and operational
performance of service firms. +/-
H4b There is a direct relationship between firm age and financial
performance of service firms. +/-
H5a There is a positive relationship between the use of ABC and
operational performance of service firms. +
H5b There is a positive relationship between the use of ABC and
financial performance of service firms. +
H6a There is a direct relationship between firm size and operational
performance of service firms. +/-
H6b There is a direct relationship between firm size and financial
performance of service firms. +/-
H7 There is a direct positive relation between differentiation strategy
and the implementation of LTPs. +
H8 There is a direct positive relation between cost leadership strategy
and the implementation of LTPs. +
H9 The differentiation strategy is positively related to the adoption of
ABC +
H10 The cost leadership strategy is negatively related to the adoption of
ABC. -
H11 There is a positive relation between the use of ABC and the use of
LTPs. +
96
Chapter 4 : Research Methodology
4.1 Introduction
In the previous chapter, the conceptual framework with associated hypotheses
representing the theory of this research has been developed. This chapter sets the
foundation for the empirical analysis by identifying the appropriate methodology that
should be adopted to examine the theory developed in this research and achieve the
overall aim of this thesis. However, identifying the appropriate methodology requires a
proper understanding of the different research paradigms, approaches, strategies and
methods available for researchers to choose from.
This chapter is divided into 12 sections. A brief but insightful review of research
paradigms, approaches and strategies is provided in sections 2, 3 and 4 respectively.
Section 5 presents the different data collection methods along with their advantages and
disadvantages to justify the selected method in this research. In section 6 the context of
this research is determined alongside the rationale behind choosing this specific context to
host the current study. Sections 7 and 8 discuss the techniques used to identify the
population and the sample from which the empirical data is going to be collected for
testing statistically the previously developed sets of hypotheses. The stages of developing
the questionnaire instrument and measures of the variables included are reported in details
in sections 9 and 10 respectively. Section 11 will determine the statistical techniques that
will be used to test the research hypotheses. Finally, section 12 concludes this chapter by
summarising its content.
4.2 Research Paradigms
The research paradigm can be thought of as the philosophical view and assumptions of a
researcher which determine how research should be conducted (Collis and Hussy, 2009).
Research paradigms can be presented on a continuum with two extremes, namely
positivism and interpretivism (Bryman and Bell, 2007). Adopting either of these two
paradigms is argued to have a significant implication on the research approach and method
97
employed for achieving the purpose of a research study (Collis and Hussy, 2009). Table 4-
1 lists the assumptions of the two main paradigms as presented in Collis and Hussy (2009,
P.58). Features of the two main paradigms are presented in Table 4-2, taken from Collis
and Hussy (2009, P.62).
Table 4-1: The assumptions of the positivism and interpretivism paradigms
Philosophical Assumption Positivism Interpretivism
Ontological Assumption
(The nature of reality)
Reality is objective and
singular, regardless of the
researcher stance
Reality is subjective and
multiple, as observed by
participants
Epistemological Assumption
(What constitutes valid
knowledge)
Researcher is independent of
what is being researched
Researcher interacts with
that being researched
Axiological Assumption
(The role of values)
Research is value-free and
unbiased
Research is value-laden and
biases are present
Rhetorical Assumption
(The language of research)
The writing is formal with
passive voice and use of
accepted quantitative words.
The writing is informal with
personal voice and accepted
qualitative words.
Methodological Assumption
(The process of research)
Process is deductive.
Study of cause and effect with
a static design (categories are
isolated beforehand).
Research is context free.
Generalisations lead to
prediction, explanation and
understanding.
findings are reliable and
precise through validity and
reliability
Process is inductive.
Study of mutual
contemporaneous shaping of
factors with an emerging
design (categories are
identified during the
process).
Research is context bound.
Patterns and/or theories are
developed for understanding.
Findings are reliable and
precise through verification. Source: Collis and Hussy (2009, P. 58)
98
Table 4-2: The main features of the positivism and interpretivism paradigms
4.2.1 Positivism
Positivism is “an epistemological position that advocates the application of the methods of
the natural sciences to the study of social reality and beyond” (Bryman and Bell, 2007:
p.16). This paradigm assumes that the existence of social reality is not dependent on us or
if we are aware of it (Collis and Hussy, 2009). Positivism is based on some principles,
among which are that knowledge is the facts gathered which provide the foundation for
laws and can be confirmed by the senses. In addition, from positivistic perspective, the
aim of theory is to produce testable hypotheses (Bryman and Bell, 2007; Saunders et al.,
2009). Positivism may involve aspects of both inductive and deductive approaches
(Bryman and Bell, 2007). However, through its practical implementation, positivism relies
more on the deductive approach with quantitative methods rather than the inductive
approach with qualitative methods (Collis and Hussy, 2009; Saunders et al., 2009).
4.2.2 Interpretivism
Interpretivism is the extreme opposite position on a spectrum including positivism and
interpretivism at its two ends (Collis and Hussy, 2009). Proponents of interpretivism
acknowledge the differences between subjects of the social and natural sciences where
Positivistic paradigm Interpretivistic paradigm
Most likely to produce quantitative data Most likely to produce qualitative data
Employs large samples Employs small samples
Focuses on hypotheses testing Focuses on hypotheses and theory generation
Data is highly specific and precise Data is rich and subjective
The location is artificial The location is natural
Data reliability is high Data reliability is low
Validity is low Validity is high
Generalises to population from sample Generalises from one setting to another
Source: Collis and Hussy (2009, P. 62)
99
studying them necessitates the reliance on different logics of research to reflect these
differences (Bryman and Bell, 2007; Collis and Hussy, 2009). Unlike positivism,
interpretivism assumes social reality to exist within us and consequently an examination
of that reality will have an effect on it (Collis and Hussy, 2009). As a result, interpretivism
predicates that differences between people and objects of the natural sciences should be
acknowledged and respected which requires social scientists to attain the subjective
meaning of social actions (Saunders et al., 2009).
4.2.3 Rationale for adopting the positivist paradigm
Given the different assumptions underlying the positivism and interpretivism paradigms,
the positivism paradigm is considered more relevant and in line with the assumptions
made in this current research. First, in line with the ontological assumption of reality in the
positivism paradigm (Collis and Hussy, 2009), it is assumed in this research that the
reality of a phenomenon like the lean-performance association or the lean-context
association exists regardless of our awareness of its existence. Second, in line with the
epistemological assumption (Bryman and Bell, 2007), it is also assumed that investigating
a phenomenon like the lean-performance association or the lean-context association by
researchers will not have any effect on the presence of those relations. Third, as stated by
Bryman and Bell (2007) and Collis and Hussy (2009), the freedom of a researcher to
decide on what research paradigm to use is constrained by the need of the paradigm to
dovetail with the nature of his/her research and the research problem. Consequently, given
the aim of this research to empirically validate the theoretical framework and hypotheses
proposed, the positivism paradigm is considered more consistent and supportive for
achieving this aim (Saunders et al., 2003).
4.3 Research approach (Deductive versus inductive approach)
After deciding on the research paradigm to be adopted, the researcher needs to make
another important decision in regard with the research approach to be used (Saunders et
al., 2009). There are two main research approaches usually used by researchers, namely
the deductive approach and the inductive approach (Bryman and Bell, 2007; Collis and
Hussy, 2009).
100
4.3.1 The deductive approach
In the deductive research, a researcher starts with information known about a specific
phenomenon to develop his/her research hypotheses that will be subject to empirical
examination in the next stage of a research project (Bryman and Bell, 2007; Saunders et
al., 2009). With the positivism paradigm being adopted, a researcher usually relies on the
available literature to develop a theory and hypotheses that need to be verified using
appropriate statistical analyses (Collis and Hussy, 2009). In other words, the deductive
research moves from the general to the specific (Collis and Hussy, 2009). Six stages are
normally subsumed in the deductive approach: (1) theory, (2) hypothesis, (3) data
collection, (4) findings, (5) hypotheses confirmed or rejected, and (6) revision of theory.
This process is depicted in Figure 4-1below.
Figure 4-1: The six stages of the deductive approach
Source: Bryman and Bell (2007, P.11)
4.3.2 The inductive approach
In contrast to the deductive approach previously explained, a researcher adopting the
inductive approach starts with real observations on a specific phenomenon and uses
Theory
Hypothesis
Data collection
Findings
Hypotheses confirmed or rejected
Revision of theory
101
his/her findings to generate theory (Bryman and Bell, 2007). In simple words, the
inductive approach follows the following sequence: observations/findings theory,
which means moving from the particular to the general (Collis and Hussy, 2009). The
main differences between the deductive and inductive approaches are presented in Table
4-3 adopted from Saunders et al. (2009, P. 127).
Table 4-3: The main differences between the deductive and inductive approaches
Deduction approach Induction approach
Scientific principles Gaining understanding of the meanings
humans attach to events
Moving from theory to data A close understanding of the research
context
The need to explain causal relationships
between variables
The collection of qualitative data
The collection of quantitative data A more flexible structure to permit changes
of research emphasis as the research
progresses
The applications of controls to ensure
validity of data
A realisation the researcher is part of the
research process
The operationalisation of concepts to
ensure clarity of definitions
Less concern with the need to generalise
A highly structured approach
Researcher independence of what is being
researched
The necessity to select samples of
sufficient size in order to generalise
conclusions
Source: Saunders et al. (2009, P. 127)
102
4.3.3 Rationale for adopting the deductive approach
Collis and Hussy (2009) emphasise the need for adopting a research approach that
supports achievement of the research aim and objectives. This research develops a
theoretical model with associated hypotheses based on the existing literature with the aim
to measure and provide empirical verification of their validity. Therefore, given the focus
and nature of the current research, the deductive approach that emphasises measurement
and empirical examination of theories and relationships between variables seems more
relevant than the inductive approach (Bryman and Bell, 2007; Saunders et al., 2009). In
addition, this approach supports more the use of large samples to improve generalisation
of results (Saunders et al., 2009) which provides an additional reason for its adoption in
this study.
4.4 Research strategy
The research strategy is the general plan of how the research questions are going to be
answered (Bryman and Bell, 2007; Saunders et al., 2009). The research strategy is usually
influenced by the research paradigm chosen (Collis and Hussy, 2009). Collis and Hussy
(2009, P.74) and Saunders et al. (2003) classify the various research strategies in terms of
their appropriateness to the two main paradigms discussed in subsection 4.2 as presented
in Table 4-4.
Table 4-4: Research strategies under the two main research paradigms
Positivism Interpretivism
Cross-sectional studies Action research
Experimental studies Case studies
Longitudinal studies Ethnography
Surveys Feminist perspective
Grounded theory
Hermeneutics
Participative enquiry
103
Given the adoption of the positivism paradigm in this study, the focus will be on research
strategies listed in the first column of the above table. I will start by excluding those
research strategies that are believed to be irrelevant given the settings and nature of the
current research.
Experimental studies can be conducted either in a laboratory or a natural setting in a
systematic manner (Collis and Hussy, 2009). This type of research strategy allows for
detecting causal relations because the researcher will have control over all variables
included in the study and can manipulate the independent variable to observe the impact
on the dependent variable (Saunders et al., 2009). However, experimental studies can be
argued to be irrelevant to this research for different reasons. First, the unit of analysis in
this research will be the firm which is difficult to be influenced (Bryman and Bell, 2007).
Therefore, it is not easy to arrange experiments in business research (Bryman and Bell,
2007; Collis and Hussy, 2009). Second, the researcher cannot have control over or
manipulate the variables included in the study which renders this research strategy
irrelevant to this current research (Bryman and Bell, 2007; Collis and Hussy, 2009).
Longitudinal studies are concerned with studying a variable or group of subjects over a
period of time (Collis and Hussy, 2009). This will permit detection of any change or
development of the relations or behaviours under examination (Bryman and Bell, 2007;
Saunders et al., 2009). However, this strategy is very time and resource consuming (Collis
and Hussy, 2009). Furthermore, the data needed for the empirical analysis of this research
is mainly internal data that can be obtained from knowledgeable members of firm
management. As a result, persuading such busy people to provide information on the same
variables more than one time can be troublesome and may bring about the inability to
obtain enough sample size to allow for generalisation of findings.
In regard to the cross-sectional and survey strategies, they are believed to be appropriate
for addressing the research problem and questions of the current study. As defined by
Bryman and Bell (2007, P. 55), Cross-sectional studies “entail the collection of data on
more than one case (usually quite a lot more than one) and at a single point in time in
104
order to collect a body of quantitative or quantifiable data in connection with two or more
variables, which are then examined to detect patterns of associations”. This strategy is
usually combined with the survey strategy to collect a large quantity of data from a large
population in a very economical way (Saunders et al., 2009). By adopting a combination
of these two research strategies, a large number of firms can be sought for participation in
this study which helps in producing more generalisable results (Bryman and Bell, 2007).
Furthermore, participants will be asked to provide the required information only once
which may prevent reducing the already decreasing response rate (Collis and Hussy, 2009;
Saunders et al., 2009).
4.5 Data collection method
Different data collection methods can be used to collect the data required for this research
given the adoption of the survey research strategy. Such methods may include interview
and questionnaire. Determining the most appropriate and efficient method is always
dependent on the advantages and disadvantages of each method and the aim and
objectives of the research. This research aims at conducting a large-scale empirical
examination of a theoretical model highlighting the lean-performance and lean-context
associations in service firms to provide generalisable results which overcomes limitations
of the previous literature. Most of the data needed to achieve the research aim and
objectives are internal (e.g. data on lean practices and MAS, etc.) and not available from
any external sources. This sort of data is mainly available in the mind of knowledgeable
management members. The two data collection methods revealing relevancy of each
method to this research are briefly reviewed below.
4.5.1 Interview method
The interview method can have different forms such as unstructured, semi-structured and
structured interview. The first two forms are mostly used to probe deeply a phenomenon
and build theory so that they are more relevant to the inductive approach not adopted in
this study (Collis and Hussy, 2009). Moreover, these methods are very time consuming
and expensive (Saunders et al., 2009). In addition, with these methods, there is also the
issue of the interviewer bias resulting from the possible variability in the way interviews
105
are conducted and questions are asked (Collis and Hussy, 2009). In contrast, the structured
interview in which questions are pre-developed and closed is suggested for the deductive
approach adopted in this study (Collis and Hussy, 2009). Therefore, this method can be
useful to obtain the necessary data for this research. However, because of the intention to
survey a large number of firms, this method seems very costly in terms of both time and
financial resources (Saunders et al., 2009). Furthermore, this method may not be
convenient to participants who have a daily busy schedule in addition to the possible
interviewer bias just explained (Collis and Hussy, 2009; Clarke and Mullins, 2001).
4.5.2 Questionnaire method
Collis and Hussy, (2009, P. 191) define a questionnaire as “a list of carefully structured
questions, chosen after considerable testing, with a view to eliciting reliable responses
from a chosen sample”. Questionnaires are a common method for data collection in
business research and there is more than one method to distribute questionnaires to sample
participants (Collis and Hussy, 2009). Such methods can be by post, online, telephone or
face-to-face (Saunders et al., 2009).
In the face-to-face method, the questionnaire can be presented at any place and time
convenient to the participant (Saunders et al., 2009). Consequently, this method becomes
very expensive and time consuming when the aim is to survey a large sample
geographically dispersed as is the case of this research (Collis and Hussy, 2009).
Therefore, this method will not be employed in this research.
The telephone method of distributing questionnaires has the advantage of being able to
survey a large sample at low cost (Saunders et al., 2009). However, like the interview
method, the telephone method introduces the issue of personal contact with its possible
related bias and may restrict the sample to those who accept to respond in this way (Collis
and Hussy, 2009). As a result, this method is also not employed in this research.
The online method seems more attractive as it reduces cost, speeds the process of
distributing questionnaires, and allows for targeting a large sample (Saunders et al., 2009).
106
In this method, a questionnaire is constructed and can be sent via email to a large number
of specified participants. However, this method requires availability of the email address
of the targeted participants. When such email addresses are not available as the case for
this study, such method cannot be relied on.
The self-administered questionnaire is the main adopted method for data collection in the
current study. This method has been chosen as it allows for surveying a large number of
firms using their addresses available easily in most databases (Clarke and Mullins 2001;
Shields, 1995). In addition, this method is widely known, economical in terms of time and
resources, more convenient to participants as they have freedom in regards to when to fill
in the questionnaire (Bryman and Bell, 2007; Collis and Hussy, 2009). However, one of
the disadvantages of this method is related to the number of questions that can be included
in a questionnaire which have a direct impact on the response rate (Collis and Hussy,
2009; Saunders et al., 2009). Therefore, archival data are also used in this research to
reduce the number of questions included in the questionnaire where this includes
information about firm age, firm size and three financial indicators as explained in section
4.10 of this chapter. Despite its disadvantage, the self-administered questionnaire is
considered the most appropriate method given the nature and overall aim of this research
and its wide use in the academic literature (e.g. Fullerton et al., 2013; Vlachos and
Bogdanovic, 2013; Banker et al., 2008; Bayo-Moriones et al., 2008; Al-Omiri and Drury,
2007; Auzair and Langfield-Smith, 2005; Shah and Ward, 2003). Figure 4-2 summarises
the methodological choices adopted in this research.
After identifying the suitable research paradigm, approach, strategy and the data collection
method, the next step is to specify the research context, population and sample that will
provide necessary data for the empirical part of this research.
4.6 Research context
The research context is an important aspect for successful theory testing in quantitative
studies (Anderson and Widener, 2007). Its importance stems from the need to ensure (i)
appropriate unit of analysis; (ii) data is available for hypotheses testing; (iii) a large
107
sample can be obtained for rigorous empirical analysis (Anderson and Widener, 2007).
Bearing in mind these considerations, the research context of this research has been
chosen to be the United Kingdom for several reasons.
Figure 4-2: Summary of the methodological choices adopted in this research
First, this research is interested in testing a theoretical model at firm level in the service
context as stated earlier. Therefore, a study with this focus should be conducted within a
country that has an established service sector where that facilitates the identification of a
suitable population and sample sufficient for data collection. Given that the UK is a
proved developed service-driven economy (Windrum and Tomlinson, 1999), it seems an
ideal context for hosting the current research.
Second, although the lean service is a recent concept that might not have been adopted by
service firms in some developed countries, lean service in the UK sounds to have been
experienced to some extent by not only the UK healthcare sector (not included in this
research), but also other service sectors as indicated by the empirical work of Alsamdi et
al. (2012). This initial indication of the familiarity of UK service firms with lean service
Research paradigm
• Positivism paradigm
Research appraoch
• Deductive approach
Research strategy
• Cross-sectional survey strategy
Data collection method
• Postal questionnaire
• Archival data
108
secures obtaining the required information for the empirical testing of the theoretical
model.
Third, during the period of conducting the current research, the researcher was resident in
the UK. As a result, contacting service firms within the UK and administration of the
postal questionnaire instrument could be easier and saved time and financial resources in
comparison with other research contexts than the UK.
4.7 Research population
A population is “the universe of units from which the sample is to be selected” (Bryman
and Bell, 2007, p. 182). The term units in this definition may indicate to people, nations,
cities, firms, etc. Given the research context determined above, the research population of
this study includes medium and large for-profit service firms (> 50 employees) located in
the UK and that have their data on age, turnover and number of employees available on
the FAME database for three years ending in 2011 included. The reason for restricting the
population of this study to only medium and large service firms is that small firms (<50
employees) are less likely to have the required resources or needs to invest in innovative
management and accounting systems such as lean service and ABC (Abdel-Kader and
Luther, 2008; Bayo-Moriones et al., 2008). Consequently, such firms may not be able to
provide relevant data needed for the empirical analysis of this research. In addition,
archival data on age and size (number of employees) is required to reduce the number of
questions asked to participants and not to rely solely on perceptual measures.
4.8 Research sampling
Identifying a sample of a population is critical for almost all quantitative studies (Collis
and Hussy, 2009). A sample is “the segment of population that is selected for
investigation” (Bryman and Bell, 2007, p. 182). The need to sample stems from the
inability of researchers in most cases to survey the whole population due to reasons such
as time and resources limitations (Bryman and Bell, 2007). In such cases, a representative
sample needs to be identified and used in the empirical analysis where the results of this
109
analysis will be generalised to the whole population from which the sample was drawn
(Collis and Hussy, 2009). There are usually two main methods for sampling (1)
probability sampling and (2) non-probability sampling (Bryman and Bell, 2007).
A probability sample is “a sample that has been selected using random selection so that
each unit in the population has a known chance of being selected” (Bryman and Bell,
2007, P. 182). Although this type of sampling does not eliminate completely sampling
errors, it helps in reducing them to the absolute minimum (Bryman and Bell, 2007).
Furthermore, adopting probability sampling permits the use of tests of statistical
significance that allow for inferences to be made about the population from which the
sample has been drawn (Collis and Hussy, 2009). Different methods can be used to ensure
that probability sampling is followed such as simple random, stratified random, systematic
and multi-stage cluster (Bryman and Bell, 2007).
In contrast to the probability sample, a non-probability sample is “a sample that has not
been selected using a random selection method” (Bryman and Bell, 2007, P. 182). In fact,
this implies that some units in the population are more likely to be selected than others.
Consequently, relying on non-probability samples yields results less generalisable to the
population in comparison with those obtained from probability samples (Bryman and Bell,
2007). Like probability sampling, different methods can be used to identify a non-
probability sample such as convenience sampling, snowball sampling and quota sampling
(Bryman and Bell, 2007).
Given the aim of conducting tests of statistical significance and making inferences about
the population, probability sampling is used in this study. More specifically, stratified
random sampling is employed rather than other types of probability sampling as this
method can ensure that the distribution of firms in the sample identified will resemble that
of the population in terms of one or more criterion (Bryman and Bell, 2007; Collis and
Hussy, 2009). The criterion used in this study to form strata is the industry type.
110
Having determined the population and sampling technique that is going to be used, the
sampling frame should be identified. Sampling frame is “the listing of all units in the
population from which the sample will be selected” (Bryman and Bell, 2007: p. 182).
Obtaining an adequate and precise sampling frame is critical in quantitative studies to
prevent sampling bias and generate more generalisable results (Collis and Hussy, 2009).
In the current study, the Financial Analysis Made Easy database 4 (FAME) has been used
to obtain the listing of all service firms in the population. This database has been chosen as
it provides detailed information on a large number of companies registered in the UK. The
sampling frame has included all medium and large (>50 employees) for-profit service
firms that provided sufficient information on its turnover and number of employees in the
three years between 2009 and 2011. In addition, the same database has been relied on to
attain details on key informants (management board members) including their names,
titles, phone numbers and addresses.
4.9 Questionnaire development
Designing a questionnaire instrument requires a large amount of care if it is to yield a
satisfactory response rate along with reliable and valid information (Collis and Hussy,
2009). This is because data can, in most cases, be collected from respondents only once
(Bryman and Bell, 2007). However, there are some critical recommendations that help to
design a more user-friendly questionnaire that allows for obtaining a high response rate
along with reliable and valid data. Among these recommendations can be creating an
attractive layout of the questionnaire, maintaining the questionnaire as short as possible,
providing clear instructions for answering the questionnaire, combining the questionnaire
with a covering letter personally addressed to each respondent and a pre-stamped return
envelope (Bryman and Bell, 2007).
Consequently, a special care in this study has been given to the design process of the
questionnaire in light of the above recommendations to produce as a user-friendly
questionnaire as possible which allows for attaining sufficient, reliable and valid data for
4 FAME database contains financial information on over 8 million private and public companies operating in
UK and Ireland. It also provides contact information for management board members of those companies.
111
conducting a rigorous empirical analysis. The final version of the questionnaire in this
study consists of three sections with 10 questions distributed over three one-sided A4
pages. The length of the questionnaire concurs with the acceptable range of lengths
provided by Saunders et al. (2009) which are between 4 to 8 A4 pages. The first section of
the questionnaire involved four questions focusing on lean service practices and their
expected benefits. The second and third section involved 2 and 4 questions respectively
that focused on contextual variables and some demographic information. The full and
final version of the questionnaire is presented in Appendix 1.
4.9.1 Questions type and format
In a questionnaire, questions can be either closed or open questions (Collis and Hussy,
2009). However, most researchers have advocated more reliance on closed questions for
positivistic studies (Collis and Hussy, 2009), which allows for choosing from a
predetermined list of answers, as they are easier and quicker to answer given of course the
salience of the topic to the participants (Bryman and Bell, 2007). Based on that, all
questions included in the questionnaire have been closed questions hoping to improve the
response rate given the length of the questionnaire. However, different response formats
can be used to generate closed questions (Collis and Hussy, 2009), but mostly rating
questions (e.g. Likert-type questions) and categorical questions have been used in the
questionnaire of the current study. All rating questions have adopted the six-point type
format which allows participants more options to express their opinion on the aspect
presented in each question.
4.9.2 Questionnaire layout and questions flow
Questionnaire layout can be critical for two main reasons: (1) to reduce non response rate
and (2) to avoid response errors (Dillman, 2007). The questionnaire layout is
recommended to be attractive so that participants are more inclined to fully fill in the
questionnaire (Collis and Hussy, 2009; Saunders et al., 2009). For these reasons, the
questionnaire has been printed out with high quality papers (A4) to make it more likable
and attractive to the participants (Dillman, 2007). Moreover, there has been an attempt to
avoid having a cramped-like questionnaire in terms of the page margins and space
112
separating the different questions (Bryman and Bell, 2007). In addition to that, the matrix
style for rating questions has been employed to save spaces (Saunders et al., 2009) and
grid line format has been used to ease the process of following questions to readers.
Furthermore, given the importance of the questions flow within a questionnaire, the
recommendation of Bryman and Bell (2007) to start the questionnaire with the most
important questions presented with logical order has been relied on.
4.9.3 Covering letter
Although the importance of covering letter accompanying the questionnaire to the
response rate has not been firmly established (Bryman and Bell, 2007), a considerable
amount of attention has been given to its content in the current study. The content of the
covering letter has revolved around providing clear and sufficient information on the
purpose and importance of the research, the reasons for which the specific participant has
been contacted and the significant impact of his/her participation, the confidentiality of the
information provided by the respondent, some instructions for answering the questions,
contact details of the researcher in the case any information or clarification has been
needed. As a way of improving response rate, each covering letter was personally
addressed to the targeted person and a promise has been made to all participants to receive
a copy of the research results as recommended by Dillman (2007). A copy of the covering
letter is presented in Appendix 2.
4.9.4 Questionnaire pre-testing process
Piloting a questionnaire instrument is an invariably essential step especially when
considering the fact that the data can be collected only once from participants (Bryman
and Bell, 2007; Collis and Hussy, 2009). The importance of this step also stems from its
expected benefits in (1) highlighting problems in the readability of the questionnaire, (2)
exposing insufficient or unclear instructions to answer the questionnaire, (3) pointing to
limitations that question the comprehension of the questionnaire to adequately cover the
topic it is intended to cover, (4) helping to identify troubling items and questions which
make respondents uncomfortable, and (5) providing a great opportunity to have
suggestions on including, dropping, or modifying some items to improve the flow, content
113
and understanding of the questions (Bryman and Bell, 2007). According to Bryman and
Bell (2007), to achieve the expected benefits of the pilot study, it is desirable to identify a
small group of participants who resemble to participants in the population from which the
sample for the full study will be drawn. As a result, the questionnaire of this study was
tested by 15 professionals from the UK service sector who are members of the Lean
Business System Group on LinkedIn. This group of participants was considered
appropriate given their expertise in the service operations and knowledge in the lean
system. All participants in the pilot study were asked to fill in the questionnaire and
provide constructive feedback in connection with clarity, readability, validity of items,
layout and flow of questions, and the need to add and/or delete items of the questionnaire.
Important feedback was received from participants in this pilot study. This included
suggestions to reposition some questions to improve the layout and flow of questions
which makes it easier to fill in the questionnaire. Participants appreciated the
comprehension of the questionnaire. All comments received were seriously acted on and
the questionnaire was modified accordingly to have its final version as presented in
Appendix 1.
4.9.5 Questionnaire administration
After amending the questionnaire instrument based on the feedback received from the
pilot study, the questionnaire was ready to be administered to the whole sample to collect
data for the main empirical analysis. Initially, 1000 questionnaires were printed out to
target the pre-identified 1000 UK for-profit service companies. Despatching
questionnaires to the targeted participants started on 10 October 2012.
A questionnaire was addressed personally to operations managers/directors when possible.
Otherwise it was personalised to other positions including chairman, CEO, managing
director or director in the management board. It was believed that people in charge of such
positions are able to provide valid and comprehensive information requested in the
questionnaire. Respondents were advised to share questions with knowledgeable persons
in their firm if they feel that would enhance accuracy and validity of information. In
majority of the cases, the targeted person completed the questionnaire. Although
114
collecting information from more than one informant per firm is desirable, this method is
expected to have a detrimental effect on the response rate (Harris, 2001). In addition,
relying on a single informant per firm is very common in the literature (e.g. Askarany et
al., 2010; Amoako-Gyampah and Acquaah, 2008; Maiga and Jacobs, 2008; Al-Omiri and
Drury, 2007).
A respondent received a copy of the questionnaire, pre-stamped envelope and introductory
letter (Appendix 2) highlighting the aim of research, providing instructions for responding
and making a clear promise to share the research results with all interested respondents. In
addition, to interpret LTPs consistently, a glossary sheet (Appendix 3) was developed
based on literature to provide participants with definition of the 23 LTPs. A reminder
letter was sent to all non-respondents almost three weeks after sending the questionnaire.
Finally telephone calls were made to encourage remaining non-respondents to participate.
Out of this process, 70 questionnaires were returned due to wrong address and 186
questionnaires were received from respondents giving a response rate of 20%5. Out of the
186 questionnaires received, 81 were returned empty for several reasons highlighted in
Table 4-5. The most frequent reasons for declining to participate were “lack of time (21
firms)” “the intended person is no longer available (20 firms)” “The questionnaire does
not apply to their industry (18 firms)”. Table 4-6 reveals that although the literature of
lean service strongly supports applicability of lean practices to all service industries, 18
firms (22%, 18/81) indicate irrelevancy of the questionnaire to their firms.
Of the 105 remaining questionnaires, six were not useable due to a large amount of
missing data leaving a final sample of 99 questionnaires. The low response rate can be
attributed to the length of questionnaire and sensitivity of information required in addition
to seeking responses from top management within the targeted organisations. However, it
is still comparable with other rates recently obtained in this field: 7.9% by Inman et al.
(2011), 10.6% by Kim et al. (2012) and 14.9% by Auzair and Langfield-Smith (2005).
5 Returning questionnaires due to wrong address is very common in survey studies. For example, among the studies which reported
this information, 469 questionnaires out of 1973 were returned due to wrong address in the study of Kroes and Ghosh (2010), 136 out of 817 in the study of Rexhausen et al. (2012), and 102 out of 1000 in the study of Al-Omiri and Drury (2007).
115
Table 4-6 provides information on our sample distribution which indicates coverage to a
wide range of service industries. The mean (median) general experience of respondents
are 18 (17) years and 9 (6) years at their current firm respectively. This provides initial
evidence on the credibility of data collected in this study.
Table 4-5: Reasons for non-participation
Reason Total
No time 21
The intended person is no longer available 20
The questionnaire does not apply to their industry 18
Company policy 9
Small company 8
Confidential information 3
High demand for participation in research studies 2
Total 81
Table 4-6: The sample distribution
Industry
Number of
respondents
Banks 13
Education 9
Hotels & restaurants 16
Insurance companies 7
Other services 24
Post and Telecommunications 8
Transport 6
Wholesale & retail trade 16
Total 99
4.9.5.1 Non-response bias
One important task of quantitative researchers employing self-administered questionnaire
to collect data is to ensure that respondents to the questionnaire do not differ from those
who decline to participate in their research (Collis and Hussy, 2009). This is known as
checking for non-response bias. Simply put, if some members of the research sample do
not participate in the research and they are different from who have participated in the
research, the research findings cannot be generalised to the population from which the
sample has been identified (Bryman and Bell, 2007).
116
Non-response bias has been tested in two different ways. First, Non-response bias has
been checked using ANOVA test on age and turnover as commonly used variables for this
purpose (e.g. De Leeuw and Berg, 2011; Craighead et al., 2009). Given the difference in
size between the sample respondents (99) and non-respondents, a random sample of 99
non-respondents was identified to be used for testing non-response bias (Hair et al., 2010;
Tabachnick and fidell, 2007). The results of ANOVA have indicated equality in means on
each variable (P-value (turnover) = 0.47, P-value (age) = 0.79) which implies non-response bias
does not seriously threaten the validity of findings in this study. The second method used
for testing non-response bias has been the wave method suggested by Armstrong and
Overton's (1977) and used largely in the literature (e.g. Grafton et al., 2010; Abdel-Kader
and Luther, 2008; Al-Omiri and Drury, 2007; Auzair and Langfield-Smith, 2005; Chong,
1996). According to this method, late respondents are expected to resemble non
respondents. Consequently, data provided by early respondents can be compared to data
from late respondents to examine non-response bias. Comparing early and late
respondents in terms of industry (χ2 = 3.75, P-value = .81) and accounting system used (χ
2
= .88, P-value = .35) has indicated no significant difference. Similarly, ANOVA has been
used to test for differences between early and late respondents on all items measuring lean
service and business strategy. The results of these tests have been positive suggesting that
non-response bias does not seriously threaten the validity of findings in this study.
4.10 Variables measurement
The decision on how to measure research variables is one of the most critical and
influential decisions researchers must make. This is so due to its direct effect on the
reliability and validity of data collected and consequently on the results achieved.
Therefore, Bryman and Bell (2007) advocate the use of existing measures whenever
possible especially if those measures have already been piloted and their reliability and
validity have been established. As a result, in the current study, the available literature for
existing measures of the variables included in this research has been thoroughly searched
so that such measures could be adopted or adapted in line of the research aim and
objectives.
117
4.10.1 Lean service measurement
As indicated before, the lean service concept is very recent and little empirical research
has been conducted in this area. Consequently, LTPs and LSPs presented in Table 3-1 and
Table 3-2 were obtained mainly from conceptual and case studies. Because of that a
conservative approach has been adopted to construct the final list of practices to be
measured and the number of references reported in the associated tables has been used as a
criterion. Specifically, for a practice to be included in the questionnaire it should be
mentioned by at least five researchers. By adhering to this criterion, 14 LTPs practices
printed bold in Table 3-1 have not been included in the questionnaire for measurement and
empirical analysis. Similarly, 7 LSPs practices printed bold in Table 3-2 have not been
included in the questionnaire for measurement and empirical analysis.
The remaining 23 LTPs and 10 LSPs presented in Table 3-1 and Table 3-2 respectively
were measured on six-point Likert type scale adapted from the existing literature. The use
of the 23 LTPs (Table 3-1) was measured using the measurement scale developed by
Fullerton et al. (2003) with slight modification. Specifically, the first point in this
measurement scale was “no consideration” in the original study, but in the current study it
was changed to “no implementation” based on feedback from the pilot study. Respondents
were asked to indicate the extent to which each of the 23 LTPs has been implemented in
their firm. Possible answers for this question were (1) no implementation, (2) considering,
(3) beginning, (4) partially, (5) substantially and (6) fully. LSPs were measured using the
measurement scale developed by Yasin et al. (2003) with slight modification regarding the
scales points to ensure consistency with the previous scale. Specifically, the original scale
was based on ten points but in the current study only six points were used. To measure
LSPs (Table 3-2), respondents were asked to indicate the level of effort spent on each of
the 10 LSPs (anchors 1= no effort to 6 = highest level of effort).
4.10.2 Organisational performance measurement
In a similar approach to that adopted for LTPs and LSPs explained in the subsection
4.10.1, all benefits presented in Table 3-3 and cited by less than 5 articles have not been
118
included in the questionnaire for measurement and empirical analysis. All six benefits
printed in bold in Table 3-3 have been dropped. The remaining 14 performance indicators
in Table 3-3 were separated into financial dimension (i.e. profitability) and operational
dimension (i.e. the other 13 indicators). To measure operational performance, the
measurement scale used by Yasin et al., 2003 was employed. Specifically, Respondents
were requested to indicate the level to which lean service practices have been effective in
delivering each of the 13 operational benefits as shown in Table 3-3 (answers: 1= strongly
disagree to 6= strongly agree).
The financial dimension of firm performance was measured by secondary data obtained
from the FAME database. Secondary data can be defined as the data not collected directly
by the researcher (Bryman and Bell, 2007: p. 326). Such data can be in different forms
such as published summaries collected by other organisations, or archival data stored in
databases like FAME, UK Nexus, and DataStream. Secondary data can be of high
importance to and will be primarily used by researchers who are interested in conducting
empirical comparison studies at national or international level (Saunders et al., 2009).
Compared to primary data, secondary data can be obtained at low cost and less time
(Bryman and Bell, 2007). Having said that does not mean that secondary data is free of
disadvantages. Researchers who use secondary data often face some difficulties regarding
the complexity and unfamiliarity with such data in addition to the lack of control over the
quality of data. However, researchers often rely on secondary data to cross validate the
primary data they have collected and consequently, both primary and secondary data are
used to answer their research questions (Saunders et al., 2009). In this research, secondary
data will be collected for two reasons: (1) to reduce the numbers of questions in the
questionnaire instrument in order to improve the response rate, and (2) to cross-validate
the findings obtained through analysing primary data collected via questionnaire
instrument (Swink and Jacobs, 2012; González-Benito, 2005). More specifically,
secondary data will be collected on firm size, firm age and three performance measures,
namely profit margin, turnover per employee, and return on capital employed.
119
Three different measures of financial performance highly used in the literature were used
which included profit margin (PM), turnover per employee (TE) and return on capital
employed (ROCE) (e.g. Agarwal et al., 2013; Bhasin, 2012; Davidson and Stern; 2004;
Patterson et al., 2004). Secondary data has been always perceived as a more reliable
source of data compared to the perceptual measures of performance. Therefore, by
combining both perceptual and secondary performance data, the findings of this study is
expected to be more robust in comparison to other studies which relied only on perceptual
measures of performance (Swink and Jacobs, 2012; González-Benito, 2005).
However, to avoid entirely arbitrary measurement of performance, the initiation year of
lean service provided by respondents has been relied on. 38 respondents answered the
question regarding the initiation year and their responses indicated that on average lean
service was implemented in 2009. This is not surprising given that lean service is a
recently developed concept (Malmbrandt and Åhlström, 2013). The year 2009 has been
perceived as the best estimate available for the implementation year of lean service based
on which financial data can be collected6. Therefore, data on the three performance
indicators have been collected on all available years since 2009 (i.e. 2009-11 included) for
all companies in the sample. The median value has then been calculated for each firm
included in the study for each financial indicator. However, because the values of these
indicators differ across industries and the sample in this study includes a wide range of
industries, an industry-adjusted median value for each firm has been calculated by
subtracting the industry median value from a firm median value (Shafer and Moeller,
2012; Swink and Jacobs, 2012; Patterson et al., 2004, Ittner et al., 2002).
In addition, current objective financial measures are expected to be correlated with past
objective financial measures, and therefore controlling for the effect of past performance
is necessary (Swink and Jacobs, 2012; Hillman, 2005; Patterson et al., 2004). Therefore,
data on PE and TE and ROCE from the preceding three years (i.e. 2006-08 included) have
6 Ensuring that the financial data is collected after the introduction of an innovation method helps in partially addressing the issue of reverse causality (Gues et al., 2003; Luft and Shields, 2003).
120
been obtained and an industry-adjusted median value has been computed for each firm and
used to control for the effect of past performance7.
4.10.3 Contextual variables measurement
Age and size were measured by objective data obtained from FAME database. Age was
measured by the number of years since inception. Size was measured by the average total
number of employees of the last three years for which data was available (2009-11). The
measurement scale for business strategy was adopted from Auzair and Langfield-Smith
(2005) in which cost leadership and differentiation strategies were measured separately
using 4 and 7 items respectively. Low values on either scale indicated low emphasis on
the corresponding strategy while high values represented the reverse. In consistence with
previous studies (e.g. Pavlatos, 2010; Banker et al., 2008; Gosselin, 1997), respondents
have been asked to indicate the type of accounting system used in their firms with the
following possible categories: variable costing, absorption costing, ABC, and others. All
categories other than ABC have been combined to represent TAS. Then a dummy variable
has been created in which 1 indicates the use of ABC and 0 otherwise (i.e. TAS). This
dummy variable has been used in the main analysis.
4.10.4 Common method bias
Given that one informant provided all data, this could raise the issue of common method
bias. Common method bias was tested statistically using Harman’s single-factor test
(Podsakoff et al., 2003). Thus, all items (LTPs, LSPs, outcome; and business strategy)
were subjected to factor analysis and the unrotated solution was examined. The result
indicates that 15 factors can be extracted with eigenvalue >1 where the first factor
explains only 19% of the total variance. This result implies that single-source bias is of
little concern in this study.
7 Controlling for past performance also assists in accounting for the possibility that past performance may explain the implementation of lean service (Gues et al., 2003).
121
4.11 Statistical techniques
In addition to the rigor required in developing a theoretical model, researchers have
stressed the need to accompany this with rigorous methodologies and statistical analyses
to verify and test the proposed theoretical model (Shields and Shields, 1998; Ittner and
Larcker, 2001; Chenhall, 2003). One of the most important statistical techniques which
has been argued to offer the level of rigor desired for theory development and testing is
Structural Equation Modelling (SEM) (Hair et al., 2010).
4.11.1 Structural Equation Modelling
SEM was first used in marketing research in the early 1980s while limited use of this
technique was observed in other fields such as operations management and management
accounting (Shah and Goldstein, 2006; Smith and Langfield-Smith, 2004). However,
given the valuable advantages of SEM, its increasing level of use has been witnessed in
different fields (Hair et al., 2011).
SEM can be thought of as “a set of multivariate techniques that allow for the simultaneous
study of the relationship between directly observable and/or unmeasured latent variables,
while incorporating potential measurement errors” (Henri, 2007: p. 76). It is proved
particularly useful for models in which a dependent variable in one equation becomes an
independent variable in another subsequent equation (Hair et al., 2010). In addition, SEM
has a superior power over other statistical techniques such as multiple regression analysis,
factor analysis, canonical correlation and path analysis in that it can test simultaneously
both the measurement properties and the theoretical relations of models (Hair et al., 2010).
Further, SEM overcomes limitations of traditional techniques (e.g. multiple regression and
path analysis) which assume an error-free measurement of constructs by explicitly
accounting for measurement errors in the estimation process (Hair et al., 2010). Based on
this description, SEM is perceived to be a three-in-one technique that combines the
features of multiple regression, factor analysis and path analysis (Hair et al., 2010).
122
However, it is crucial to point out that SEM can be regarded as a family of techniques
(Peng and Lai, 2012; Hair et al., 2011; Smith and Langfield- Smith, 2004), which
encompass covariance-based SEM (CB-SEM) and variance-based SEM or Partial Least
Squares SEM (PLS-SEM). In the following section, the main characteristics and
differences between the two streams will be discussed.
4.11.2 Comparison between CB-SEM and PLS-SEM
When the term SEM is used, many academics think of CB-SEM that is usually
implemented by well-known software like AMOS and LISREL (Chin, 1998; Hair et al.,
2011). However, SEM is not limited to CB-SEM and it includes another distinctive but
relatively less popular technique, namely PLS-SEM (Peng and Lai, 2012).
PLS-SEM is “a causal modelling approach aimed at maximising the explained variance of
the dependent latent constructs” (Hair et al., 2011, P. 139). Consequently, it differs from
CB-SEM’s objective which mainly concerns reproducing the theoretical covariance
matrix, without focusing on explained variance (Hair et al., 2011). Specifically, CB-SEM
attempts to produce a set of parameter estimates which minimise the differences between
the estimated covariance matrix and the sample covariance matrix (Hair et al., 2011). In
contrast, PLS-SEM attempts to estimate a set of parameter estimates that minimise the
residual variances of latent dependent constructs (Hair et al., 2012).
Apart from the objective of the two techniques, CB-SEM model estimation requires a
number of assumptions to be fulfilled such as the multivariate normality of data, number
of indicators per construct and large sample size (Hair et al., 2011). When these
assumptions cannot be met, and they are usually difficult to be met in business research
(Peng and Lai, 2012), CB-SEM may not result in precise estimation parameters (Hair et
al., 2011; Hair et al., 2010). In such cases, PLS-SEM can be preferred given its capability
of handling small sample sizes and not normally distributed data in addition to allowing
the use of constructs with only one or two indicators (observed variables) (Hair et al.,
2011; Hair et al., 2012; Peng and Lai, 2012; Lee et al., 2011). However, when the
assumptions of CB-SEM are met, CB-SEM is preferred as it produces more precise
123
parameter estimates (Hair et al., 2010). One more difference between PLS-SEM and CB-
SEM is that the latter produces goodness of fit indices necessary for theory testing while
the former does not. Therefore, when the research objective is theory testing and
confirmation, CB-SEM is more appropriate, while PLS-SEM is more suitable for theory
development (Peng and Lai, 2012; Hair et al., 2011; Hair et al., 2010). Form the above
discussion it can be seen the advantages of PLS-SEM are the disadvantages of CB-SEM
and vice versa, and consequently the two approaches are complementary rather than
competitive (Hair et al., 2012). Therefore, researchers should choose the technique that
best suits their research objective, data characteristics and model complexity (Hair et al.,
2012). To help researchers in choosing between the two approaches, Hair et al. (2011, P.
143) state that “when CB-SEM assumptions are violated with regard to normality of
distributions, minimum sample size, and maximum model complexity, or when related
methodological matters emerge.... PLS-SEM is a good methodological alternative for
theory testing”.
In the current study, PLS-SEM is adopted for three reasons. First, the subjective and
objective data collected somewhat violate the normality assumption necessary for CB-
SEM. Second, the model proposed is complex given the number of constructs included
and the number and nature of structural relations hypothesised. Third, the sample size in
this research is relatively small.
4.11.3 Overview of PLS-SEM
Like CB-SEM, PLS-SEM has two components for testing latent variable models, namely
measurement (outer) model and structural (inner) model (Hair et al., 2011). The
measurement model associate observed indicators to their respective latent variables,
while the structural model links endogenous latent variables to exogenous latent variables
(Hair et al., 2010). Endogenous latent variables represent dependent variables and
exogenous latent variables are similar to independent variables in multiple regression
analysis (Hair et al., 2010). PLS-SEM can handle two types of measurement model,
reflective and formative models (Hair et al., 2011). Reflective models accommodate
reflective constructs where a change in the latent construct leads to change in the
124
associated indicators. In contrast, formative model encompass constructs where a change
in the observed indicators of a construct leads to change in the construct (Hair et al.,
2010). However, all constructs in the current research are reflective and the procedures
appropriate for evaluating the reflective measurement models are followed in this study.
To estimate the proposed model, the PLS-SEM algorithm uses a two-stage approach (Hair
et al., 2011). The latent variables’ scores are estimated in the first stage while the final
outer weights and loadings are calculated in the second stage (Hair et al., 2011).
4.11.3.1 Assessing the Measurement Model under PLS-SEM
The first step in assessing PLS-SEM is to evaluate to the measurement model in terms of
its measures’ reliability and validity. The measures’ reliability can be assessed through
different criteria such as internal consistency (composite reliability) and indicator
reliability (Hair et al., 2011). Construct validity is usually evaluated through convergent
and discriminant validity (Hair et al., 2010). More details on testing constructs reliability
and validity will be presented later in the next chapter.
4.11.3.2 Assessing the Structural Model under PLS-SEM
As mentioned before, PLS-SEM seeks to estimate model parameters that maximise the
variance of the dependent latent constructs explained by the latent independent constructs.
Therefore, R2
and path coefficient along with their significance should be the primary
assessment criteria (Hair et al., 2011). Because PLS-SEM relaxes the distribution
assumption of the data, the significance of path coefficients are evaluated using re-
sampling techniques such as bootstrapping or jackknifing (Hair et al., 2011). However,
bootstrapping is usually perceived to be superior to the jackknifing method (Chin, 1998)
and therefore it is adopted in this study. Another important criterion for assessing the
structural model is its predictive capability (Hair et al., 2011). The predictive capability of
the model is usually evaluated by the Stone-Geisser Q2 value (Geisser, 1974; Stone,
1974), which assumes that the model must be able to predict each endogenous latent
construct’s indicator (Hair et al., 2011). This Q2
value is calculated using the blindfolding
technique which omits part of the data systematically and uses the resulting estimates to
predict the omitted part of the data (Hair et al., 2011).
125
4.11.3.3 PLS-SEM Software
PLS-SEM algorithms were developed few decades ago although their use in business
research has not been observed until recently (Hair et al., 2011). This limited use of PLS-
SEM can in part be attributed to the slow progress in developing user friendly, yet
rigorous, software for rigorous PLS-SEM analysis in business and management research
(Temme et al., 2010). However, these days there are different software packages that
support PLS-SEM such as PLS-GUI, Visual-PLS, PLS-Graph, Smart PLS, SPAD-PLS
(Temme et al., 2010). Although all the aforementioned software packages support PLS-
SEM, each has some distinctive features in terms of options available. In this study,
Smart-PLS will be used as it offers all necessary options to evaluate a PLS model. First,
compared to other packages Smart-PLS is a free-to-use package that can be downloaded
from Smart-PLS community website. Second, Smart-PLS allows for automatically
building product terms for interaction and moderation analysis. Third, this package has the
blindfolding feature required to evaluate the predictive relevance of the model. On the
other hand, one limitation of Smart-PLS package is that it does not offer P-value to assess
the significance of path coefficients. However, it does provide t-statistics through the
bootstrapping procedure which can be used to assess the significance of path coefficients.
4.12 Summary of chapter four
In this chapter the research methodology adopted in this research study has been discussed
in details. By doing so, the research paradigm, approach and strategy were all identified
and justified after exploring the other alternatives adopted in social sciences. More
specifically, this study adopted the positivism paradigm with a deductive approach and the
cross-sectional survey strategy to test the theoretical model and associated hypotheses
developed in this study. For data collection, the postal questionnaire was found to be the
ideal option given the setting and aim of the research. Consequently, the measurement
scales of variables were determined and the questionnaire instrument was developed,
piloted and refined. Moreover, this chapter discussed the research context, population and
the required sample for providing empirical data. The administration process of the
questionnaire was presented along with some descriptive statistics on respondents and the
126
test of non-response bias and common method bias. The results of these tests indicated
that non-response bias and common method bias should not be of high concern to the
current study. Finally, a description of the statistical techniques to be used in this study
was provided along with the rationale behind their adoption.
127
Chapter 5 : Data preparation and examination
5.1 Introduction
This chapter is devoted mainly to examine and prepare the data collected for conducting
the empirical analysis through PLS-SEM. The second section of this chapter explicates the
procedures adopted for screening the empirical data collected through the questionnaire
instrument. This includes a detailed discussion of the missing data analysis and the known
assumption of parametric tests including normality, linearity, and homogeneity of
variance, outliers and multicollinearity. The third section of this chapter revolves around
verifying the validity, reliability and unidimensionality of the constructs subsumed in the
theoretical model of this study. After establishing the validity, reliability and
unidimensionality of constructs, the descriptive statistics of those constructs including
mean, standard deviation, skewness and kurtosis are reported in section four.
5.2 Preliminary screening of data
5.2.1 Missing data analysis
Missing values happen when a respondent does not provide his/her answer to one or more
of the survey questions. As a result, valid values for those questions will be missing from
the analysis, which necessitates evaluation of the pattern and extent of the missing data, to
probe the reasons behind the missing data (Hair et al., 2010).
A proper understanding of the reasons behind the missing data assists in determining the
optimal method to deal with it (Hair et al., 2010). As per Hair et al. (2010), there can be
two types of missing data. First, ignorable missing data (<10%) is that anticipated given
the research design and the technique used. This type of missing data does not need
specific remedies. Second, non-ignorable missing data (>10%) is the one that occurs due
to some procedural factors or factors pertinent to the respondents. This type of missing
data necessitates finding appropriate remedies. To identify the type of missing data in a
data set, Hair et al. (2010) recommends examining the extent and patterns of the missing
data.
128
Examining the extent of missing data in this research study using the SPSS package shows
that missing values (<2%) are within the ignorable level (10%) of missing data for all
variables. Furthermore, the Little’s MCAR test has been relied upon to examine the
pattern of missing data. This test investigates whether the missing data have any
systematic pattern or they are missing completely at random (Hair et al., 2010; Tabachnick
and Fidell, 2007). The results of Little’s MCAR test (Chi-square 127.112, p = 0.27)
indicate that the data does not suffer from any systematic error, which allows for higher
flexibility in choosing the method for treating the missing data (Hair et al., 2010). As a
result, this study adopts the mean substitution method for missing values, as one of the
most common ways to compute missing values (Tabachnick and Fidell, 2007; Hair et al.,
2010).
5.2.2 Examining the assumptions of parametric tests
To the extent that one or more of the parametric tests will be used to examine the research
hypotheses, the first step after the quantitative data has been collected is then to ensure its
validity in connection with the assumptions of parametric tests (Hair et al., 2010). This
includes examining the raw data for its normality, linearity, homogeneity of variance and
outliers (Hair et al., 2010). The skewness and kurtosis of the data were used to examine its
normality while linearity and homogeneity of variance were checked using the scatter plot
graph as recommended by Hair et al. (2010). The Z-score of higher than 3.3 was relied on
to identify potential outliers in the data set (Hair et al., 2010).
Given their effect on other assumptions, the screening process started by attempting to
identify the potential outliers in the data set. Observations with Z score larger than 3.3
were winsorised to avoid reducing the sample size. When examining the normality
assumption, 6 LTPs (italic in Table 3-1) were found to have a significantly skewed
distribution (P<.001). Their distributions could not be improved by different
transformations and consequently they were deleted from further analysis. In terms of the
linearity and homogeneity of variance assumptions, scatter plots of a large sample of pairs
of variables indicated no apparent violation to these assumptions. Consequently, the
129
screening process ended after ensuring that 17 LTPs (Table 3-1), 10 LSPs (Table 3-2) and
13 operational performance indicators (Table 3-3) were valid input for the factor analysis
(Hair et al., 2010).
5.3 Validity, reliability and unidimensionality of constructs
Examination of the validity, reliability and unidimensionality of measures used in a
quantitative study is an essential task (Hair et al., 2010; Bryman and Bell, 2007). This is
so because each of these aspects has a significant influence on the extent to which a
researcher can be confident about his/her findings resulted from the use of statistical tests
to examine the research hypotheses (Hair et al., 2010). More specifically, validity and
reliability of measures provide indications on the level of measurement error present in a
data set which is a direct influential factor on the credibility of a research’s findings.
5.3.1 Content validity
Content validity, known also as face validity, concerns the assessment of the
appropriateness of variables used to represent a specific concept in really reflecting the
content and the theoretical definition of that concept (Bryman and Bell, 2007). This type
of validity is usually examined subjectively through seeking feedback from experts on the
concepts being investigated or pre-tests with some subpopulations (Hair et al., 2010).
Ensuring the content validity of constructs in this study was achieved through a relatively
large-scale pilot study that sought feedback from 15 experts in the concepts measured in
this study (see subsection 4.9.4).
5.3.2 Unidimensionality of constructs
The unidimensionality of a construct implies that all indicators of that construct are
strongly correlated with each other and represent only that specific construct (Hair et al.,
2010). The importance of establishing the unidimensionality of all constructs in a model
stems from the confusion that may arise when some indicators represent more than one
construct. Factor analysis, such as exploratory factor analysis (EFA) and confirmatory
factor analysis (CFA), is a critical statistical tool usually used to empirically establish the
unidimensionality of constructs. The test using the factor analysis technique revolves
130
around making sure that all indicators of a factor load highly and significantly only on that
factor (Hair et al., 2010).
In this study, EFA is employed to empirically assess the dimensionality of constructs
included in the model using the SPSS software package. In this analysis the principal
component method with varimax rotation and eigenvalue greater than 1 was used as a
criterion for factor extraction. The principal component method was chosen because it
takes into consideration the total variance including common, specific and error variances
(Hair et al., 2010). Moreover, the varimax orthogonal rotation method was adopted as in
most cases the un-rotated solutions are neither sufficient nor clear (Hair et al., 2010). In
addition, the orthogonal rotation methods have been relied on more widely in comparison
to oblique rotation methods (Hair et al., 2010).
The recommendations given by Hair et al. (2010) for a reliable factor analysis were
followed for all constructs. Kaiser-Meyer- Olkin (KMO) test for sampling adequacy was
used at a scale and individual item level with a minimum value of 50% being acceptable.
All items in a scale should have communalities of at least 50% and their loadings should
be >= 55% given the sample size of about 100 observations (Hair et al., 2010). Any item
that did not satisfy these conditions was removed from the analysis and a new factor
solution was requested again until all items had satisfied the required conditions.
5.3.3 Reliability of constructs
Reliability is an assessment of the consistency of a measure of a concept (Bryman and
Bell, 2007, P. 163). There are two different forms or meanings of the term reliability,
namely stability also called test-retest method and internal reliability (Hair et al., 2010).
5.3.3.1 The stability method
The stability method concerns the consistency of a measure in providing almost similar
results at two different points in time (Bryman and Bell, 2007). By adopting this method, a
measure should be administered to a sample on one occasion and then re-administered to
the same sample on another occasion. The measure is said to be reliable if the results
131
obtained from the two administrations are highly correlated (Hair et al., 2010). However,
this method has some limitations which render it inapplicable to this research. First, in the
case of questionnaire data, a respondent’s answers at time 1 may influence his/her answers
at time 2 (Bryman and Bell, 2007). Second, this method obviously requires administration
of the same measures twice to the same respondents which is expensive, time consuming
and needs convincing participants to provide the same information twice.
5.3.3.2 The internal reliability method
This method applies to multiple-indicator constructs where data collected on all indicators
is aggregated to make an overall score for the associated construct (Hair et al., 2010). The
objective of this method is to ensure that the multiple indicators which measure a specific
construct are related to each other because they all measure the same construct (Bryman
and Bell, 2007). Cronbach’s alpha is one of the most used techniques for testing the
internal reliability of multiple-indicator constructs when factor analysis is used (Hair et al.,
2010; Bryman and Bell, 2007). In general, there is an agreement among researchers that a
Cronbach’s alpha value of 70% should be the lowest acceptable limit for this coefficient
(Hair et al., 2010). However, given the sensitivity of this coefficient to the number of
indicators in a construct (i.e. the value of Cronbach’s alpha increases with the increase in
the number of indicators used in measuring a construct, even with the same degree of
inter-correlation), a coefficient value of 60% (Hair et al., 2010) or 50% (Nunnally, 1978)
can be acceptable especially in exploratory research or for constructs with low number of
indicators (Hair et al., 2010; Grafton et al., 2010; Cortina, 1993).
A second measure of internal reliability is the composite reliability measure (Hair et al.,
2011). Unlike Cronbach’ alpha, the composite reliability measure does not suppose that all
indicators are equally reliable and this makes it more appropriate for PLS-SEM used in
this study (Hair et al., 2011). Satisfactory reliability can be assumed when the value of the
composite reliability ranges between 0.7 and 0.9 for advanced research while a value of
0.6 is considered acceptable for exploratory studies (Hair et al., 2011).
132
In the current research, both Cronbach’s alpha and composite reliability measure will be
checked to ensure that the measures used in this study are satisfactorily reliable.
5.3.4 Validity of constructs
After ensuring the unidimensionality, reliability of a construct and its measures, the final
step is to examine the construct validity (Hair et al., 2010). Construct validity can be
defined as “the extent to which a scale or set of measures accurately represent the concept
of interest” (Hair et al., 2010: p. 126), or “the issue of whether or not an indicator (or set
of indicators) that is devised to gauge a concept really measures that concept” (Bryman
and Bell, 2007, P. 165). In addition to the content validity explained earlier in subsection
5.3.1, two other forms of construct validity can be measured and empirically tested,
namely convergent validity and discriminant validity (Hair et al., 2010).
5.3.4.1 Convergent Validity
Convergent validity concerns the evaluation of the extent to which indicators of a specific
construct converge or share a high amount of variance in common (Hair et al., 2010).
Convergent validity can be empirically assessed by different ways among which can be
the reliance on factor loadings or the average variance extracted (AVE) when performing
CFA (Hair et al., 2010). When using the factor loadings method to assess the convergent
validity of a construct, the standardised loading of each indicator measuring that construct
should be examined. A standardised loading value of 0.5 and ideally 0.7 can be an
indication of convergent validity (Hair et al., 2010). Regarding the second method, the
AVE can be calculated as the sum of all squared standardised factor loadings divided by
the number of items. An AVE value of 0.5 or higher can indicate a good level of
convergent validity (Hair et al., 2010). This is because when the AVE value falls below
0.5, it implies that, on average, the items variance explained by the latent factor is less
than the error variance remains without explanation.
5.3.4.2 Discriminant Validity
Discriminant validity assesses the level to which each construct is distinct from other
constructs (Hair et al., 2010). Like convergent validity, discriminant validity can be
133
empirically examined by two methods, namely the cross-loadings method and AVE
method. By adopting the cross-loadings method, discriminant validity can be supported by
ensuring higher loadings of indicators on their specified construct in comparison with their
loadings on other constructs (Hair et al., 2010; Chin, 1998). However, the AVE method,
usually accompanying CFA, to assess the discriminant validity requires that AVE for any
two constructs to exceed the squared value of the correlation estimate between these two
constructs (Hair et al., 2010).
5.3.5 Empirical examination of unidimensionality, reliability and validity of
constructs
As explained in the previous subsections of this chapter, EFA is implemented to assess the
validity and unidimensionality of the model constructs which will be confirmed by the
measurement model of PLS-SEM used in this study.
5.3.5.1 Assessing unidimensionality, reliability and validity of LTPs
After deleting 6 LTPs because they violate some of the assumptions of parametric tests
(see subsection 5.2.2), the remaining 17 LTPs presented in standard format in Table 3-1
were subjected to EFA. Table 5-1 presents the results of the factor analysis of the 17
LTPs. As can be seen from Table 5-1, four factors were extracted that explained 62% of
the data variance. However, one practice “Mistake proofing” was dropped as it had no
significant loading (>55) on any factor. The factor solution presented in Table 5-1
confirmed the unidimensionality of each factor extracted. All indicators related to a
specific factor were loading significantly (>55%) on only that factor with values ranging
from 61% to 83%. In addition, no high cross loadings were evident. Therefore, the results
of Table 5-1 also confirmed the convergent and discriminant validity of the four factors
extracted. Kaiser’s measure of sampling adequacy (0.78) indicated that EFA is appropriate
and within acceptable levels (Hair et al., 2010).
The reliability of each factor was estimated using the Cronbach’s alpha. As shown in
Table 5-1, all factors possessed a satisfactory reliability value ranging from 0.68 to 0.83.
Based on the indicators (i.e. LTPs) loaded on each factor, the four factors were labelled as
134
process factor, physical structure factor, customer value factor and error prevention factor,
respectively.
Table 5-1: Factor analysis and reliability analysis of lean technical practices
Practices Factor loadings Communality
1 2 3 4
Automation 0.702 0.170 0.003 -0.178 0.554
Just in Time 0.620 0.441 0.006 -0.118 0.593
Pull system 0.760 0.024 0.175 0.306 0.703
Work load balancing 0.731 0.066 0.231 0.179 0.624
Quick set up time 0.708 0.067 0.256 0.296 0.659
Small lots 0.643 0.352 0.158 -0.209 0.606
5Ss 0.063 0.706 0.146 0.189 0.560
Group technology 0.231 0.768 0.211 0.090 0.696
Improving facility layout 0.177 0.820 0.080 0.217 0.757
Visualisation 0.166 0.607 0.354 0.052 0.524
Kaizen blitz 0.402 0.135 0.607 0.096 0.557
Policy deployment/Hoshin Kanri 0.155 0.098 0.799 0.109 0.684
Quality function deployment 0.167 0.201 0.697 0.200 0.594
Value stream mapping -0.025 0.308 0.598 -0.272 0.527
Root cause analysis -0.026 0.202 0.167 0.736 0.612
Total preventive maintenance 0.127 0.177 -0.020 0.831 0.738
Measure of sampling adequacy (Whole model) 0.777
Variance extracted by the model 62.417
Cronbach's alpha 0.832 0.81 0.711 0.677
5.3.5.2 Assessing unidimensionality, reliability and validity of LSPs
Table 5-2 presents the results of factor analysis of the 10 LSPs listed in Table 3-2. As can
be seen from Table 5-2, two factors were extracted that explained 75% of the data
variance. However, one practice (multifunctional employees) was dropped due to low
communality value. The factor solution presented in Table 5-2 confirmed the
unidimensionality of each factor extracted. All indicators related to a specific factor were
loading significantly (>55%) on only that factor with values ranging from 82% to 89%. In
addition, no high cross loadings were evident. Therefore, the results of Table 5-2 also
confirmed the convergent and discriminant validity of the two factors extracted. Kaiser’s
135
measure of sampling adequacy (0.85) indicated that EFA is appropriate and within
acceptable levels (Hair et al., 2010).
The reliability of each factor was estimated using the Cronbach’s alpha. As shown in
Table 5-2, all factors possessed a satisfactory reliability value ranging from 0.90 to 0.91.
Based on the indicators (i.e. LSPs) loaded on each factor, the two factors were labelled as
motivation factor and human factor, respectively.
Table 5-2: Factor analysis and reliability analysis of lean social practices
Practices Factor loadings Communality
1 2
Reward system 0.837 0.064 0.705
Communication system 0.845 0.121 0.729
Management support 0.867 -0.046 0.753
Performance measures 0.875 0.111 0.779
Training 0.822 0.167 0.704
Employee empowerment 0.034 0.873 0.763
Employee commitment 0.066 0.886 0.789
Employee involvement 0.120 0.887 0.801
Leadership 0.117 0.853 0.742
Measure of sampling adequacy (Whole model) 0.848
Variance extracted by the model 75.156
Cronbach's alpha 0.907 0.902
5.3.5.3 Assessing unidimensionality, reliability and validity of organisational
performance
The 13 operational performance indicators presented in Table 3-3 were factor analysed.
As can be seen from Table 5-3, three operational performance factors were extracted
explaining 68% of the data variance. However, two operational indicators “reduction in
inventory and improvement in capacity” were dropped because of low sampling adequacy
(<50%) and low communality value (<50%) respectively. The factor solution presented in
Table 5-3 confirmed the unidimensionality of each factor extracted. All indicators related
to a specific factor were loading significantly (>55%) on only that factor with values
ranging from 67% to 85%. In addition, no high cross loadings were evident. Therefore, the
136
results of Table 5-3 also confirmed the convergent and discriminant validity of the three
factors extracted. Kaiser’s measure of sampling adequacy (0.82) indicated that EFA is
appropriate and within acceptable levels (Hair et al., 2010).
The reliability of each factor was estimated using the Cronbach’s alpha. As shown in
Table 5-3, all factors possessed a satisfactory reliability value ranging from 0.71 to 0.83.
Based on the indicators loaded on each factor, the three factors were labelled as customer
satisfaction, waste elimination and process time reduction, respectively.
Table 5-3: Factor analysis and reliability analysis of operational performance
Practices Factor loadings Communality
1 2 3
Customer perception of product/service quality 0.789 0.290 0.193 0.745
Customer satisfaction 0.848 0.232 0.051 0.776
Employees satisfaction and their performance 0.834 0.132 0.229 0.765
Employees understanding of the process 0.709 0.248 0.274 0.639
Identification and elimination of waste 0.108 0.668 0.323 0.562
Operational efficiency 0.394 0.703 0.095 0.658
Productivity 0.250 0.712 0.276 0.645
Reduction in costs 0.182 0.812 -0.071 0.697
Freeing staff time 0.309 0.225 0.742 0.696
Reduction in lead time and cycle time 0.077 -0.002 0.814 0.668
Human errors 0.211 0.228 0.781 0.707
Measure of sampling adequacy (Whole model) 0.823
Variance extracted by the model 68.712
Cronbach's alpha 0.828 0.708 0.768
5.3.5.4 Assessing unidimensionality, reliability and validity of business strategy
Indicators of the cost leadership and differentiation strategy were factor analysed
separately. Table 5-4 and Table 5-5 summarise the results of the factor analysis for these
scales. As shown in Table 5-4, cost leadership items loaded on one factor explaining 63%
of the variance. Similarly, items measuring differentiation strategy loaded on one factor
explaining 69% of the variance as indicated in Table 5-5. The factor solution for both
scales confirmed the unidimensionality of the two factors extracted. All indicators loaded
137
significantly (>55%) on their related factor with values ranging from 73% to 85% and
75% to 88% for cost leadership and differentiation respectively. Therefore, the results also
confirmed the convergent and discriminant validity of this construct. Kaiser’s measure of
sampling adequacy was 0.72 (0.89) for cost leadership (differentiation) models which
indicated that EFA is appropriate and within acceptable levels (Hair et al., 2010). The
alpha level was 81% and 93% for cost leadership and differentiation respectively which
provided support to the respective scales.
5.4 Descriptive statistics
This section presents the descriptive statistics of variables measured in the questionnaire
and objectively collected from the FAME database. These descriptive statistics are
important as they provide initial view of the nature of the data used in the main statistical
analysis (Tabachnick and Fidell, 2007). The mean, standard deviation, minimum value,
maximum value, skewness and kurtosis of LTPs, LSPs, organisational performance,
business strategy, MAS, firm size, firm age, unionisation, and internationalisation are
reported below.
5.4.1 Descriptive statistics of LTPs
As found in subsection 5.3.5.1, four factors reflected the technical side of lean service.
Table 5-6 conveys the descriptive statistics of these dimensions along with their associated
indicators. First, by examining the average score (printed in bold) of the four factors of the
technical side, Table 5-6 indicates that on average these factors have not widely been
implemented by the service firms included in the sample.
The average score of each factor is just slightly above the average score of the scale (3 out
of 6). Among the four factors, error prevention factor and customer value factor have, on
average, the highest scores of 3.73 and 3.24 respectively. This can be a positive sign in
that it implies that service firms in the sample understand the core of lean service which is
to improve the value delivered to customers. As indicated by Womack and Jones (1996)
the first principle of lean service is to understand the value from customer perspective.
138
Table 5-4: Factor analysis and reliability analysis for cost leader strategy
Practices Factor loadings Communality
1
Achieving lower cost of services than competitors 0.727 0.528
Making service/procedures more cost efficient 0.845 0.714
Improving the cost required for coordination of various services 0.811 0.659
Improving the utilisation of available equipment, services and facilities 0.794 0.630
Measure of sampling adequacy (Whole model) 0.722
Variance extracted by the model 63.272
Cronbach's alpha 0.805
Table 5-5: Factor analysis and reliability analysis for differentiation strategy
Practices Factor loadings Communality
1
Providing high quality services 0.750 0.562
Customising services to customers need 0.881 0.776
Providing after-sale services and support 0.831 0.691
Introducing new services/procedures quickly 0.856 0.733
Providing services that are distinct from that of competitors 0.848 0.719
Offering a broader range of services than the competitors 0.815 0.664
Improving the time it takes to provide services to customers 0.828 0.686
Measure of sampling adequacy (Whole model) 0.891
Variance extracted by the model 69.027
Cronbach's alpha 0.925
At individual indicator level, Table 5-6 highlights few practices that have an average score
below the average score of the scale. This includes JIT (2.84 out of 6), quick set up (2.94
out of 6), small lots (2.59 out of 6) and value stream mapping (2.86). These results support
the findings of Alsmadi et al. (2012) who also find that some lean practices (quick set up)
implemented in manufacturing to be of less relevance to services. Collectively, Table 5-6
indicates that service firms subsumed in the sample are at the early stage of implementing
LTPs and their focus on understanding the value from customer perspective can be a
positive sign for a successful implementation of the program.
139
An examination of the skewness and kurtosis of factors and individual indicators
presented in the last two columns of Table 5-6 demonstrates no serious violation of the
normality assumption (p < 001).
Table 5-6: Descriptive statistics of LTPs factors and indicators
Construct Minimum Maximum Mean Std. Deviation Skewness Kurtosis
Process factor 1 6 3.01 1.63 0.20 -1.14
Automation 1 6 3.22 1.64 -0.08 -1.32
Just in Time 1 6 2.84 1.59 0.41 -0.95
Pull system 1 6 3.20 1.68 0.11 -1.32
Work load balancing 1 6 3.28 1.43 -0.17 -0.98
Quick set up time 1 6 2.94 1.77 0.37 -1.18
Small lots 1 6 2.59 1.68 0.59 -1.11
Physical structure factor 1 6 3.14 1.66 0.00 -1.36
5Ss 1 6 3.07 1.71 0.13 -1.41
Group technology 1 6 3.04 1.56 -0.02 -1.34
Improving facility layout 1 6 3.46 1.72 -0.26 -1.39
Visualisation 1 6 2.99 1.66 0.17 -1.33
Customer value factor 1 6 3.24 1.65 -0.06 -1.25
Kaizen blitz 1 6 3.38 1.65 -0.12 -1.16
Policy deployment/Hoshin Kanri 1 6 3.34 1.69 -0.21 -1.30
Quality function deployment 1 6 3.37 1.70 -0.07 -1.33
Value stream mapping 1 6 2.86 1.58 0.14 -1.23
Error prevention factor 1 6 3.73 1.57 -0.40 -0.89
Root cause analysis 1 6 3.74 1.50 -0.32 -0.85
Total preventive maintenance 1 6 3.72 1.65 -0.48 -0.93
5.4.2 Descriptive statistics of LSPs
The social side of lean service has been reflected by two factors as shown in subsection
5.3.5.2. Unlike the four LTP factors, all LSP factors have mean values which exceed the
average value of their associated scale, that is, over 3. It seems from Table 5-7 that, on
average, UK service companies in the sample employed in this study have focused
exceptionally on the human factor (3.97 out of 6) followed by the motivation factor (3.72
out of 6). The higher focus on the social side of lean service observed in Table 5-7
emphasises earlier evidence reported by Alsamdi et al. (2012) who also notice higher
140
interest in the social practices among service firms compared to manufacturing firms. At
individual indicator level, training (4.05 out of 6) and management support (3.82 out of 6)
present themselves as the most important elements of the motivation factor. Regarding the
human factor, leadership (4.21 out of 6) and employee involvement (4.13 out of 6) seem
to be the most important elements.
An examination of the skewness and kurtosis of factors and individual indicators
presented in the last two columns of Table 5-7 demonstrates no serious violation of the
normality of the data. All factors and indicators have skewness and kurtosis values less
than 3.29 (p < .001) as recommended by Hair et al. (2010).
Table 5-7: Descriptive statistics of LSP factors and indicators
Construct Minimum Maximum Mean Std. Deviation Skewness Kurtosis
Motivation factor 1 6 3.72 1.43 -0.20 -0.84
Reward system 1 6 3.39 1.44 0.22 -0.95
Communication system 1 6 3.69 1.45 -0.25 -0.70
Management support 1 6 3.82 1.30 -0.25 -0.79
Performance measurement system 1 6 3.65 1.49 -0.30 -0.95
Training 1 6 4.05 1.45 -0.44 -0.83
Human factor 1 6 3.97 1.31 -0.33 -0.67
Employee empowerment 1 6 3.65 1.39 -0.03 -0.97
Employee commitment 1 6 3.89 1.22 -0.30 -0.72
Employee involvement 1 6 4.13 1.31 -0.36 -0.68
Leadership 1 6 4.21 1.33 -0.64 -0.32
5.4.3 Descriptive statistics of organisational performance
As mentioned before, organisational performance is measured at both operational and
financial levels. Operational performance is measured with subjective data obtained from
respondents while financial performance is measured by secondary data on three
indicators, namely profit margin (PM), return on capital employed (ROCE) and turnover
per employee (TE).
141
Table 5-8 presents the descriptive statistics of the operational performance factors and
their indicators. As can be seen, respondents seem to be optimistic about the outcome of
implementing lean service practices. The three operational performance factors have an
average score higher than the average score of the measurement scale (i.e. 3). However,
Table 5-8 reveals that, on average, the highest improvement achieved by the sample firms
of this study is in the customer satisfaction with mean value of 4.23 (out of 6). In addition,
waste elimination seems to be the second most appreciated improvement factor attributed
to lean service given its mean value of 4.10 (out of 6). This is an interesting result in that it
supports the capability of lean service to fulfil its promise in improving the customer value
by eliminating waste from processes (Ehrlich, 2006; Womack and Jones, 1996).
An examination of the skewness and kurtosis of the three factors and individual associated
indicators presented in the last two columns of Table 5-8 demonstrates no serious
violation of the normality of the data. All factors and indicators have skewness and
kurtosis values less than 3.29 (p < .001) as recommended by Hair et al. (2010).
Table 5-9 presents the descriptive statistics of the three financial indicators used in this
study. Examining the minimum, maximum and mean columns in Table 5-9 highlights a
wide variation in the three financial measures and the potential for outliers which together
render the data of these variables non-normally distributed. The data of these variables
will be cleaned as indicated in subsection 5.2.2 before using it in the main analysis.
5.4.4 Descriptive statistics of business strategy
Table 5-10 presents the descriptive statistics of the cost leadership and differentiation
strategy. As can be seen, the sample firms in this study on average emphasises the
differentiation strategy to a larger extent compared to the cost leadership strategy indicated
by the higher mean score of the former (4.10 out of 6) in comparison with the latter (3.83).
In connection with individual indicators, all indicators for both strategies have a mean
score higher than the average of the scale (over 3).
142
Table 5-8: Descriptive statistics of operational performance factors and indicators
Construct Minimum Maximum Mean Std. Deviation Skewness Kurtosis
Customer satisfaction 1 6 4.23 1.22 -0.44 -0.29
Customer perception of product/service quality 1 6 4.31 1.32 -0.57 -0.25
Customer satisfaction 1 6 4.37 1.24 -0.46 -0.55
Employees satisfaction and their performance 1 6 4.07 1.15 -0.27 -0.19
Employees understanding of the process 1 6 4.16 1.18 -0.47 -0.19
Waste elimination 1 6 4.10 1.11 -0.23 -0.22
Identification and elimination of waste 1 6 4.10 1.14 -0.20 -0.07
Operational efficiency 1 6 4.18 1.11 -0.23 -0.42
Productivity 1 6 4.07 0.96 -0.07 0.34
Reduction in costs 1 6 4.04 1.23 -0.42 -0.74
Process time reduction 1 6 3.44 1.33 0.04 -0.79
Freeing staff time 1 6 3.48 1.36 0.15 -0.82
Lead time and cycle time 1 6 3.29 1.41 0.10 -0.90
Number of human errors 1 6 3.56 1.22 -0.11 -0.64
Table 5-9: Descriptive statistics of financial performance
Construct Minimum Maximum Mean Std. Deviation Skewness Kurtosis
Profit Margin -49.70 70.71 7.36 13.45 1.44 9.94
ROCE -14.12 371.72 33.22 56.09 3.57 15.94
Turnover per employee 8,227.59 2,027,750 276,005.71 364,629.05 2.47 6.80
However, the most emphasised items in the cost leadership strategy are “making
service/procedures more cost efficient” with an average score of 4 out of 6 and
“Improving the utilisation of available equipment, services and facilities” with an average
score of 3.85 out of 6. On the other hand, the most emphasised indicators for the
differentiation strategy are “Providing high quality services” with an average score of 4.47
out of 6 and “Customising services to customers need” with an average score of 4.30 out
6. This underlines the very different focus and priorities for the two different strategies as
stated by Porter (1980).
An examination of the skewness and kurtosis of the two factors and individual associated
indicators presented in the last two columns of Table 5-10 demonstrates no serious
143
violation of the normality of the data. All factors and indicators have skewness and
kurtosis values less than 3.29 (p < .001) as recommended by Hair et al. (2010).
Table 5-10: Descriptive statistics of business strategy
Construct Minimum Maximum Mean Std. Deviation Skewness Kurtosis
Cost leadership strategy 1 6 3.83 1.31 -0.21 -0.61
Achieving lower cost of services than competitors 1 6 3.77 1.35 -0.26 -0.61
Making service/procedures more cost efficient 1 6 4.00 1.24 -0.26 -0.40
Improving the cost required for coordination of
various services
1 6 3.71 1.27 -0.13 -0.43
Improving the utilisation of available equipment,
services and facilities
1 6 3.85 1.40 -0.18 -1.02
Differentiation strategy 1 6 4.10 1.48 -0.41 -0.84
Providing high quality services 1 6 4.47 1.51 -0.74 -0.57
Customising services to customers need 1 6 4.30 1.46 -0.63 -0.59
Providing after-sale services and support 1 6 3.93 1.55 -0.30 -1.05
Introducing new services/procedures quickly 1 6 3.92 1.49 -0.18 -0.95
Providing services that are distinct from that of
competitors
1 6 4.10 1.48 -0.41 -0.95
Offering a broader range of services than the
competitors
1 6 3.84 1.54 -0.19 -1.13
Improving the time it takes to provide services to
customers
1 6 4.12 1.32 -0.42 -0.61
5.4.5 Descriptive statistics of MAS
The costing system used by service firms is measured categorically as shown in Table 5-
11. The frequency table below indicates that (37%) of service firms in the sample rely on
ABC to provide information about their processes and services. In addition, the second
most used costing system is the absorption system (33%) followed by the variable costing
system (27%).
Table 5-11: Descriptive statistics of MAS
MAS
Type of Costing System Frequency Percent
Variable costing 27 27
Full absorption costing 33 33
Activity-based costing (ABC) 37 37
Others 2 2
Total 99 100
144
5.4.6 Descriptive statistics of internationalisation and Unionisation
Table 5-12 reveals the level of internationalisation and unionisation of the sample firms
participated in this research study. As can be seen, around 27% and 61% of the sample
firms have zero level of internalisation (foreign sales) and unionisation respectively.
However, about 55% (41+14) of firms have up to 50% of their sales from foreign markets
while 32% (29+3) are unionised up to 50%.
Table 5-12: Descriptive statistics of internationalisation and unionisation
Internationalisation Unionisation
Level of internationalisation/Unionisation Frequency Percent Frequency Percent
0 27 27.27 60 60.6
1-25% 41 41.41 29 29.3
26-50% 14 14.14 3 3
51-75% 14 14.14 4 4
76-100% 3 3.03 3 3
Total 99 100 99 100
5.4.7 Descriptive statistics of firm age and size
Table 5-13 shows that the average age of firms in this study is 31 years which indicates
that more mature companies at the account of young companies are represented. The
skewness and kurtosis values are above the acceptable level of 3.29 suggested by Hair et
al. (2010) which highlights violation to the normality assumption. Similarly, the average
number of employees in this study is around one thousand which is far away from the
maximum value of twelve thousands and the minimum value of 50. Moreover, the high
skewness (3.41) and kurtosis (13.52) values assure violation of the normality assumption.
For this reason PLS-SEM has been chosen as it does not assume normality of the data
(Hair et al., 2011).
145
Table 5-13: Descriptive statistics of firm age and firm size
Indicator Minimum Maximum Mean Std. Deviation Skewness Kurtosis
Age 9.00 120.00 31.03 24.89 1.99 3.72
Number of employees 50.00 11,989.00 1,014.15 1,923.63 3.41 13.52
5.4.8 Correlations and multicollinearity
Multicollinearity takes place when independent variables in a model are strongly
associated with each other. The ideal situation for a researcher is to have a high correlation
between the independent variables and the dependent variable, but no or little correlation
between the independent variables (Hair et al., 2010). High levels of multicollinearity
negatively impact the validity of results produced by the examined model because they
effect imprecise estimation of the regression coefficients and sometimes their sign too
(Hair et al., 2010). “As multicollinearity increases, the total variance explained decreases.
Moreover, the amount of unique variance of independent variable is reduced to levels that
make estimation of their individual effects quite problematic” (Hair et al., 2010, p. 201).
One method for assessing multicollinearity is to examine the correlation matrix of
independent variables. The presence of high correlations between independent variables
(0.90 or more) can be an indication of a multicollinearity problem (Hair et al., 2010). The
correlation matrix of the IVs, reported in Table 5-14, included in this study has been
scrutinised to spot all high correlations indicating a multicollinearity problem. The
correlations presented in Table 5-14 do not indicate the presence of multicollinearity
problem given that the highest correlation is 49% which is far less than the 90% value
suggested by Hair et al. (2010). Table 5-15 presents the correlation matrix of all DVs and
IVs of this study.
Table 5-15 offers some insights into the relationships between lean service factors and
firm operational and financial performance. As can be seen from the table, all correlations
between lean service factors and performance factors are positive although not necessarily
significant. Surprisingly, none of the correlations between lean service factors and waste
elimination is significant at 5% significance level. However, a large number of significant
correlations are evident between lean service factors and the other two operational factors,
146
namely customer satisfaction and process time reduction. At financial level measured by
secondary data, Table 5-15 reveals a relatively weaker association between lean service
factors and the three financial variables.
Table 5-14: The correlation matrix of the independent variables
Independent Variables 1 2 3 4 5 6 7 8 9 10 11
1 Age 1
2 ABC -0.095 1
3 Differentiation strategy 0.116 .273** 1
4 Leadership strategy 0.101 -.261** 0.12 1
5 Motivation factor -0.182 -0.134 0.065 .272** 1
6 Human factor -0.096 -0.037 .269** 0.051 0.19 1
7 Process time factor -0.102 .201* .281** .260** 0.139 .269** 1
8 Physical structure factor -0.145 .213* 0.139 .220* .207* 0.184 .463** 1
9 Customer value factor -0.033 .202* .307** 0.172 .200* .240* .456** .491** 1
10 Error prevention factor -0.172 0.015 0.06 -0.03 0.123 0.156 0.177 .325** .219* 1
11 Size .333** -0.129 -0.14 0.096 -0.122 -0.041 -0.072 0.073 0.086 -0.079 1
**. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).
Table 5-15: The correlation matrix of all DVs and IVs of this study
Variables Customer
satisfaction
Waste
elimination
Process time
reduction PM ROCE TE
ABC 0.05 -0.09 0.08 0.06 -0.15 0.10
Differentiation strategy .230* .214* 0.15 .202* 0.04 0.05
Leadership strategy .309** .358** .304** -0.03 0.09 0.08
Motivation factor .203* 0.20 .285** .337** .211* 0.14
Human factor 0.07 0.10 0.01 0.17 .312** 0.07
Process factor .279** 0.12 .280** 0.18 0.13 0.05
Physical structure factor .296** 0.13 .314** .244* 0.05 0.06
Customer value factor .365** 0.07 .348** 0.15 0.11 0.10
Error prevention factor 0.08 0.09 .251* 0.04 0.04 0.00
Age 0.00 0.10 0.08 -0.09 -0.19 -0.07
Size 0.01 0.00 0.05 -0.13 0.03 -.232*
**. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).
147
5.5 Summary of chapter five
In this chapter, the empirical data collected through the questionnaire instrument in
addition to the secondary financial data were screened, cleaned and tested against the
assumption of parametric tests. Using EFA, the unidimensionality, validity and reliability
of measures were examined. In consequence, lean technical factors, lean social factors and
operational performance factors were established. Finally descriptive statistics of all
constructs included in this study were presented.
148
Chapter 6 : Statistical analysis using PLS-SEM
6.1 Introduction
This chapter will focus on testing the theoretical model and associated hypotheses
developed in chapter 3 of this study by a means of PLS-SEM. PLS-SEM estimates
simultaneously the measurement and structural model (Peng and Lai, 2012; Hair et al.,
2012). However, the measurement model should first be evaluated to ensure construct
reliability and validity and then the structural model representing the research hypotheses
can be examined (Hair et al., 2011; Lee et al., 2011).Therefore, in the second section of
this chapter, the measurement model linking the various constructs presented in the
previous chapter and in Figure 3-3 will be evaluated. In the third section of this chapter,
the structural model representing the set of hypotheses developed in chapter three will be
formally tested.
6.2 Measurement model
The aim of the measurement model is to ensure that all constructs employed in the model
are reliable and valid for testing the structural relations, as there is no point of testing such
relations with invalid and/or unreliable measures (Hair et al., 2011). The first step for
constructing the measurement model is to identify all constructs along with related
indicators that should be included in the measurement model. Constructs included in the
measurement model will be reported below.
6.2.1 Constructs of the measurement model
Information on the constructs of the measurement model will be obtained from the
analysis conducted in the previous chapter. First, the measurement model will include the
four factors representing the technical side of lean service along with their associated
indicators presented in Table 5-6. In addition, the model will include the two factors
representing the social side of lean service along with their associated indicators reported
in Table 5-7. Further, two constructs indicating the type of business strategy adopted
(differentiation vs. cost leadership) which has been established in Table 5-10 will also be
149
involved. As shown in Table 5-11, the costing system used by the sample firms has been
measured by a categorical measure with three categories, namely variable costing system,
absorption costing system and ABC. Therefore, the first two categories are combined to
represent companies using traditional accounting systems. A dummy variable is then
formed in which ABC is decoded 1 and 0 otherwise. Operational performance will be
represented by the three factors presented in Table 5-8 along with their associated
indicators. In regard to the financial performance, the three financial performance
measures reported in Table 5-9 will also be included. However, because objective
financial measures are usually different for different industries, these three measures have
been adjusted by subtracting the industry median performance from the performance of
the sample firms (Shafer and Moeller, 2012; Swink and Jacobs, 2012). Finally, firm size
and age as presented in Table 5-13 and three variables to control for the effect of past
performance of PM, ROCE and TE are added. Figure 6-1 depicts the measurement model.
The associated indicators of each construct have not been added for simplicity. However,
these indicators can be found in Tables 5-5 to 5-13 as has been explained above.
6.2.2 Measurement model evaluation
6.2.2.1 Construct reliability
To evaluate construct reliability in PLS-SEM, Hair et al. (2011) suggest the reliance on
composite reliability measure rather than Cronbach’s alpha as the former does not assume
all indicators are equally reliable which makes it more suitable to PLS-SEM. A construct
is said to be reliable if its composite reliability value is above 0.70 for advanced research
or above 0.60 for exploratory research (Hair et al., 2011).
However, Smartpls calculates both the composite reliability measure and Cronbach’s
alpha value for each construct. These are shown in Table 6-1. As can be seen from Table
6-1, the composite reliability value for all constructs is well beyond the acceptable value
of 0.70. In addition, the Cronbach’s alpha for all but “Error prevention factor” is higher
than 0.70. This strongly supports the reliability of constructs used in this study.
150
Figure 6-1: The measurement and structural model
6.2.2.2 Construct validity (convergent and discriminant)
Hair et al. (2011) point out that construct validity in a reflective measurement models like
the one in this study can be assessed by examining the convergent and discriminant
validity (see subsection 5.3.4). Convergent validity of a construct is evident if AVE of that
construct is 0.5 or higher (Hair et al., 2011). With regard to the discriminant validity, two
methods can be used the Fornell-Larcker criterion and cross loadings (Hair et al., 2011).
For a construct to have a discriminant validity under the Fornell-Larcker criterion (Fornell
and Larcker, 1981), AVE of that construct should be higher than the squared correlation of
that construct with any other construct in the model. This method can also be applied by
comparing the square root of AVE of a specific construct with its correlations with other
constructs in the model (Hair et al., 2010). Under the second method (i.e. cross loadings),
Differentiation Cost leadership
ABC
Past
performance
Size
Age
Error
prevention
factor
Customer
value factor
Physical
structure
factor
Process factor
Human factor
Motivation
factor
Process time
reduction
Waste
elimination
Customer
satisfaction
Turnover per
employee
ROCE
Profit
Margin
+ +
+
-
+
+
+
+
+ +
+
+/-
+/-
+/-
+/-
151
discriminant validity is evident when an indicator’s loading with its associated construct is
higher than its loading with any other construct in the model (Hair et al., 2011).
Table 6-1: Reliability and validity measures of constructs
Construct AVE Composite Reliability Cronbach’s Alpha
Age 1 1 1
ABC 1 1 1
Differentiation 0.69 0.94 0.92
Cost leadership 0.61 0.86 0.81
Motivation factor 0.73 0.93 0.91
Human factor 0.77 0.93 0.90
Process factor 0.54 0.87 0.83
Physical structure factor 0.64 0.88 0.81
Customer value factor 0.54 0.82 0.71
Error prevention factor 0.76 0.86 0.68
Customer satisfaction 0.74 0.89 0.83
Waste elimination 0.62 0.83 0.71
Process time reduction 0.68 0.86 0.77
Profit Margin 1 1 1
ROCE 1 1 1
Size 1 1 1
Turnover per employee 1 1 1
As can be seen from Table 6-1, AVE for all constructs in the model is higher than the
suggested value of 0.50 which supports the convergent validity of those constructs. To
examine the discriminant validity of constructs following the Fornell-Larcker criterion, the
correlation matrix between constructs is constructed as presented in Table 6-2. As can be
noticed from Table 6-2, the square root of AVE of each construct is higher than its
correlation with any other construct in the model which clearly supports the discriminant
validity of the constructs.
To double check the discriminant validity using the cross loadings method, Table 6-3 has
been constructed. Exploring the results in Table 6-3 reveals that all indicators have
significantly higher loadings with their respected constructs compared to their loadings
with other constructs in the model. In other words, no serious sign of cross loadings which
further supports the discriminant validity of the constructs employed in this study.
152
Table 6-2: The correlation matrix of constructs and square root of AVE
Construct 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1 Age 1.00
2 ABC -0.10 1.00
3 Differentiation 0.12 0.27 0.83*
4 Cost leadership 0.10 -0.26 0.12 0.78
5 Motivation factor -0.18 -0.13 0.07 0.27 0.85
6 Human factor -0.10 -0.04 0.27 0.05 0.19 0.88
7 Process factor -0.10 0.20 0.28 0.26 0.14 0.27 0.73
8 Physical structure factor -0.15 0.21 0.14 0.22 0.21 0.18 0.46 0.80
9 Customer value factor -0.03 0.20 0.31 0.17 0.20 0.24 0.46 0.49 0.74
10 Error prevention factor -0.17 0.01 0.06 -0.03 0.12 0.16 0.18 0.32 0.22 0.87
11 Customer satisfaction 0.00 0.06 0.23 0.31 0.20 0.08 0.28 0.30 0.38 0.08 0.86
12 Waste elimination 0.12 -0.11 0.22 0.36 0.21 0.08 0.12 0.12 0.07 0.13 0.48 0.79
13 Process time reduction 0.08 0.10 0.14 0.30 0.27 0.02 0.30 0.33 0.36 0.26 0.46 0.41 0.83
14 Profit Margin -0.09 0.06 0.20 -0.03 0.34 0.17 0.18 0.24 0.15 0.04 0.01 0.02 0.14 1.00
15 ROCE -0.19 -0.15 0.04 0.09 0.21 0.31 0.13 0.05 0.11 0.04 0.11 0.00 -0.03 0.44 1.00
16 Size 0.33 -0.13 -0.14 0.10 -0.12 -0.04 -0.07 0.07 0.09 -0.08 0.00 -0.01 0.06 -0.13 0.02 1.00
17 Turnover per employee -0.07 0.10 0.05 0.08 0.14 0.07 -0.05 0.06 0.10 0.00 0.03 0.06 0.03 -0.08 -0.13 -0.23 1.00
* Values in the diagonal represent the square root of AVE for each construct.
Having the measurement model passed the reliability and validity tests of its constructs,
the structural model now can be estimated and the hypotheses of this study be tested
accordingly (Hair et al., 2011).
6.3 Structural model
Having ensured that the constructs included in the model of this study are valid and
reliable, the analysis can proceed to estimate and evaluate the structural model
representing the hypotheses developed in chapter 3. For this, all constructs presented in
Table 6-4 should be included in the structural model. Compared to the measurement
model, the structural model has a product term of each of the four technical factors of lean
service with each of its two social factors. These are important for testing the synergy
proposed in H3a and H3b (see Table 3-4).
153
Table 6-3: Correlation matrix of constructs and indicators
Indicators 1 2 3 4 5 6 7 8 9 10 11
Differ1 0.72 0.19 0.29 0.27 -0.04 0.10 0.13 0.12 0.20 0.25 0.17
Differ2 0.88 0.01 0.28 0.21 0.08 0.28 0.06 0.05 0.25 0.21 0.02
Differ3 0.86 0.00 0.25 0.31 0.21 0.32 0.13 0.06 0.15 0.16 0.13
Differ4 0.88 0.04 0.21 0.32 0.13 0.32 0.04 0.11 0.25 0.18 0.14
Differ5 0.82 0.10 0.16 0.17 0.15 0.19 -0.04 0.25 0.20 0.30 0.17
Differ6 0.80 0.00 0.10 0.09 0.05 0.25 -0.03 0.08 0.06 0.15 0.06
Differ7 0.82 0.11 0.26 0.18 0.15 0.23 0.04 0.08 0.24 0.23 0.17
Reward system 0.01 0.84 0.14 0.09 0.14 0.23 -0.02 0.17 0.15 0.13 0.13
Communication system -0.07 0.85 0.05 0.06 0.16 0.12 0.10 0.29 0.14 0.24 0.29
Management support 0.08 0.89 0.19 0.17 0.24 0.22 0.13 0.30 0.22 0.17 0.27
Performance measurement system 0.16 0.82 0.23 0.09 0.19 0.14 0.26 0.22 0.22 0.18 0.21
Training 0.11 0.86 0.20 0.16 0.14 0.12 0.09 0.19 0.16 0.28 0.27
Employee empowerment 0.24 0.20 0.83 0.20 0.17 0.17 0.08 0.16 0.10 0.10 0.05
Employee commitment 0.21 0.13 0.88 0.29 0.18 0.24 0.09 0.04 0.10 0.09 0.01
Employee involvement 0.32 0.15 0.90 0.25 0.17 0.30 0.18 0.04 0.02 0.10 0.00
Leadership 0.19 0.20 0.90 0.21 0.14 0.12 0.19 0.00 0.03 0.08 0.02
Automation 0.19 0.13 0.23 0.75 0.25 0.25 -0.01 0.28 0.29 0.30 0.29
Just in Time 0.06 0.10 0.16 0.74 0.42 0.30 0.12 0.13 0.23 0.10 0.30
Pull system 0.33 0.05 0.17 0.74 0.32 0.35 0.28 0.24 0.09 0.08 0.14
Quick set up time 0.28 0.10 0.27 0.70 0.34 0.44 0.27 0.24 0.21 0.02 0.11
Small lots 0.16 0.00 0.14 0.73 0.41 0.36 0.03 0.12 0.23 0.11 0.24
Work load balancing 0.29 0.23 0.23 0.72 0.33 0.39 0.19 0.12 0.09 -0.08 0.15
5Ss 0.11 0.15 0.14 0.25 0.75 0.34 0.23 0.19 0.24 -0.05 0.18
Group technology 0.19 0.13 0.17 0.44 0.85 0.45 0.25 0.11 0.21 0.11 0.23
Improving facility layout 0.08 0.21 0.19 0.39 0.88 0.35 0.32 0.22 0.29 0.12 0.31
Visualisation 0.06 0.17 0.08 0.38 0.72 0.44 0.23 0.19 0.19 0.13 0.34
Kaizen blitz 0.34 0.06 0.18 0.46 0.38 0.76 0.15 0.03 0.30 -0.09 0.29
Policy deployment/Hoshin Kanri 0.21 0.19 0.19 0.30 0.36 0.80 0.17 0.10 0.24 0.07 0.31
Quality function deployment 0.21 0.28 0.30 0.36 0.40 0.77 0.29 0.23 0.34 0.15 0.27
Value stream mapping 0.12 0.04 0.01 0.18 0.31 0.58 0.00 0.16 0.15 0.05 0.20
Root cause analysis 0.08 0.17 0.21 0.12 0.27 0.23 0.88 0.01 0.11 0.13 0.21
Total preventive maintenance 0.02 0.05 0.06 0.18 0.29 0.14 0.86 -0.07 0.03 0.06 0.24
Leader1 -0.03 0.14 -0.02 0.11 0.02 -0.01 -0.07 0.59 0.11 0.13 0.13
Leader2 0.06 0.22 -0.02 0.16 0.05 -0.06 0.03 0.77 0.24 0.33 0.23
Leader3 0.15 0.26 0.03 0.30 0.27 0.19 0.02 0.88 0.27 0.42 0.28
Leader4 0.10 0.21 0.12 0.19 0.22 0.26 -0.09 0.86 0.32 0.24 0.26
Customer satisfaction 0.15 0.16 -0.09 0.14 0.16 0.14 0.04 0.30 0.80 0.47 0.30
Employees satisfaction and their performance 0.24 0.17 0.11 0.29 0.26 0.32 0.08 0.20 0.88 0.43 0.41
Employees understanding of the process 0.19 0.19 0.11 0.27 0.32 0.44 0.07 0.32 0.87 0.50 0.44
Productivity 0.22 0.15 0.10 0.16 0.14 0.07 0.10 0.38 0.47 0.80 0.40
Reduction in costs 0.14 0.15 0.11 0.05 0.11 0.06 -0.01 0.23 0.38 0.72 0.15
Identification and elimination of waste 0.15 0.19 0.01 0.05 0.05 0.04 0.17 0.25 0.38 0.77 0.36
Freeing staff time 0.14 0.30 -0.02 0.16 0.15 0.19 0.15 0.30 0.48 0.43 0.80
Lead time and cycle time 0.10 0.14 0.04 0.35 0.36 0.40 0.25 0.21 0.31 0.21 0.82
Number of human errors 0.12 0.27 0.01 0.20 0.27 0.27 0.22 0.25 0.41 0.39 0.85
(1) Differentiation, (2) Motivation factor, (3) Human factor, (4) Process factor, (5) Physical structure factor, (6) Customer value
factor, (7) Error prevention factor, (8) Cost leadership, (9) Customer satisfaction, (10) Waste elimination, (11) Process time
reduction
154
Table 6-4: Constructs included in the structural model with respective reasons for their
inclusion
Constructs Reason
Process factor H1a-H1b, H7, H8, H11
Physical structure factor H1a-H1b, H7, H8, H11
Customer value factor H1a-H1b, H7, H8, H11
Error prevention factor H1a-H1b, H7, H8, H11
Motivation factor H2a-H2b
Human factor H2a-H2b
Motivation factor * Process factor H3a-H3b
Human factor * Process factor H3a-H3b
Motivation factor * Physical structure factor H3a-H3b
Human factor * Physical structure factor H3a-H3b
Motivation factor * Customer value factor H3a-H3b
Human factor * Customer value factor H3a-H3b
Motivation factor * Error prevention factor H3a-H3b
Human factor * Error prevention factor H3a-H3b
Age H4a-H4b
ABC H5a-H5b, H9, H10, H11
Size H6a-H6b
Differentiation H7, H9
Cost leadership H8, H10
Lag profit margin To control for previous performance effect
Lag ROCE To control for previous performance effect
Lag turnover/employee To control for previous performance effect
Performance constructs
Customer satisfaction H1a-H6b
Waste elimination H1a-H6b
Process time reduction H1a-H6b
Industry-adjusted Profit margin H1a-H6b
Industry-adjusted ROCE H1a-H6b
Industry-adjusted Turnover/employee H1a-H6b
The hypotheses developed in Chapter 3 and summarised in Table 3-4 will be tested using
PLS-SEM and the Smartpls package in three stages as explained below.
6.3.1 Stages for hypotheses testing using PLS-SEM
In the first stage, the structural model will be estimated with all variables in Table 6-4
being included except the interaction terms necessary for H3a and H3b. This stage is
important for testing the main effect of the technical and social lean service factors (i.e.
H1a-H2b). As explained by Hair et al. (2013), a model that has an independent variable X,
an independent variable Z and an interaction term X*Z cannot be used to estimate the
main effect of X and Z on a dependent variable Y. The logic behind this argument is that
155
when the interaction term (X*Z) is included, the coefficients of X and Z represent their
conditional rather than main effect. That is, the coefficient of X indicates the effect of X
on Y when Z is zero. Similarly, the coefficient of Z indicates the effect of Z on Y when X
is zero (Hair et al., 2013).
In the second stage, eight interaction terms (four technical factors * two social factors,
printed in bold in Table 6-4) are created using the feature available in Smartpls for testing
H3a and H3b. The Smartpls package creates an interaction latent variable term (e.g.
Motivation factor * Process factor) by building product terms using the indicators of the
latent independent variables (i.e. Motivation factor and Process factor). These product
terms serve as indicators of the interaction term (i.e. Motivation factor * Process factor) in
the structural model (Henseler and Chin, 2010). To reduce the level of multicollinearity
usually resulting from the inclusion of interaction terms, Smartpls allows for standardising
the indicators of the latent independent variables used for creating the product terms
serving as indicators of the eight interaction terms (Henseler and Chin, 2010). However,
due to one unique feature of PLS-SEM, the coefficient of the 8 interaction terms cannot be
interpreted without adjustment (Henseler and Chin, 2010).
For an appropriate and valid interaction analysis, an interaction term (X*Z) is created after
standardising both X and Z. Although X and Z enter the analysis in their standardised
form, the interaction term (X*Z) should not be standardised (Henseler and Chin, 2010).
However, as Henseler and Chin (2010) highlight, PLS calculates path coefficients from
standardised latent variable scores. That is, for the structural model, PLS will standardise
all latent variables including the interaction terms. This renders the resulting interaction
term path coefficients invalid for interpretation without adjustment. For making these
coefficients interpretable, Henseler and Chin (2010) suggest modifying the standard
deviation of the interaction term’s latent variable score before calculating the structural
model with the interaction term. This can be done by multiplying the latent variable scores
of the interaction term by the weighted average of the standard deviations of the product
indicators using the respective loadings as weights (Henseler and Chin, 2010).
156
Having adjusted the latent variable scores of the eight interaction terms included in the
model, the third and final stage for testing H3a and H3b will begin by estimating the
structural model with all variables presented in Table 6-4 are included with the adjusted
scores of the interaction terms. It is the results of this third round estimation that will be
used to examine the interaction effect suggested by H3a and H3b.
6.3.2 Structural model specification and evaluation
The structural model of this study will be tested using Smartpls 2.0. This version performs
a PLS analysis with the following settings: the Weighting Scheme is the “Path Weighting
Scheme”, Data Metric is “Mean 0, Var1”, Maximum Iterations are “300”, Abort criterion
is “1.0E-5” and Initial Weights are “1”.
To evaluate the structural model, as indicated in subsection 4.11.3.2, R2 and path
coefficients are the primary indications (Hair et al., 2011). The significance of path
coefficients will be obtained through bootstrapping technique. Hair et al. (2011) point out
that the larger the number of samples used during the bootstrapping process, the more
robust the findings will be. Therefore, instead of relying on the default number of 200 for
bootstrapping in Smartpls 2.0, the bootstrapping process will be applied on 500 samples
with the number of cases is equal to the sample size of this study which is “99”. Another
important criterion for assessing the structural model is its predictive capability (Hair et
al., 2011). The predictive capability of the model is usually evaluated by the Stone-Geisser
Q2 value (Geisser, 1974; Stone, 1974), which assumes that the model must be able to
predict each endogenous latent construct’s indicator (Hair et al., 2011). This Q2
value is
calculated using the blindfolding technique which omits part of the data systematically
and uses the resulting estimates to predict the omitted part of the data (Hair et al., 2011).
Two forms of Q2
are usually produced by Smartpls the cross-validated communality and
the cross-validated redundancy. However, Hair et al. (2011) recommend using the latter
rather than the former as it uses the PLS-SEM estimates of both the structural and
measurement models for data prediction. A Q2
value of larger than zero implies that the
exogenous constructs have predictive relevance for the endogenous constructs included in
the model. Table 6-5 presents the two forms of Q2
.
157
Table 6-5: Cross validated Communality and Redundancy
Construct Cv-communality Cv-redundancy
Age 1.0000
Differentiation 0.6869
Lag profit margin 1.0000
Lag ROCE 1.0000
Lag turnover/employee 1.0000
Cost leadership 0.6151
Motivation factor 0.7285
Human factor 0.7693
Size 1.0000
ABC 1.0000 0.1628
Process factor 0.5423 0.0925
Physical structure factor 0.6440 0.0817
Customer value factor 0.5450 0.0745
Error prevention factor 0.7603 0.0020
Customer satisfaction 0.7300 0.3301
Waste elimination 0.6286 0.2231
Process time reduction 0.6792 0.3030
Profit margin 0.9926 0.6030
ROCE 0.9967 0.5947
Turnover/employee 0.9962 0.7953
As can be seen, all values in Table 6-5 are larger than zero indicating a satisfactory level
of predictive relevance of the model.
6.4 Hypotheses testing
In this section the research hypotheses will be tested and reported. For this purpose, the
hypotheses are classified into two groups. The first group includes hypotheses focusing
the relationship between lean service bundles, contextual variables and firm performance.
The second group includes those hypotheses that focus on the impact of business strategy
and ABC on the lean service bundles. The results are reported in Tables 6-6 and 6-7.
158
Table 6-6: Lean bundles, contextual variables and firm performance-results from PLS-SEM
Endogenous variables
ABC
Process
factor
Physical
structure
factor
Customer
value
factor
Error
prevention
factor
Customer
satisfaction
Waste
elimination
Process
time
reduction PM ROCE TE
Standardised coefficient (β)
Differentiation 0.31
(3.89)***
0.18
(1.83)*
0.03
(0.27)
0.23
(2.42)**
0.07
(0.65)
Cost leadership -0.3
(3.29)***
0.3
(2.82)***
0.29
(2.54)**
0.19
(1.59)
-0.04
(0.33)
ABC 0.23
(2.21)**
0.28
(3)***
0.19
(1.9)*
-0.02
(0.12)
-0.03
(0.33)
-0.09
(0.92)
0.06
(0.57)
0.05
(0.48)
0
(0.01)
0.09(1.36)
Motivation factor 0.12
(1.1)
0.19
(1.52)
0.24
(2.23)**
0.2
(2.32)**
0.04
(0.46)
0.01(0.12)
Human factor -0.06
(0.54)
0.01
(0.12)
0.14
(1.28)
0.14
(1.49)
0.28
(3.42)***
0.01(0.14)
Process factor 0.11
(0.9)
0.09
(0.62)
0.15
(1.14)
0.002
(0.01)
0.14
(1.58)
0.03(0.46)
Physical structure factor 0.14
(1.16)
0.08
(0.58)
0.09
(0.9)
0.06
(0.46)
-0.11
(1)
-0.05(0.81)
Customer value factor 0.27
(1.98)**
-0.04
(0.28)
0.19
(1.54)
0.06
(0.57)
0.02
(0.19)
-0.01(0.16)
Error prevention factor 0.05
(0.44)
0.11
(0.85)
0.18
(1.9)*
-0.1
(0.9)
0.01
(0.06)
0.03(0.4)
Age 0.05
(0.52)
0.2
(1.67)*
0.17
(1.53)
-0.05
(0.53)
-0.11
(1.34)
0.09(1.44)
Size -0.04
(0.39)
-0.05
(0.57)
0.04
(0.4)
-0.09
(1.04)
0.13
(1.75)*
Lag profit margin 0.54
(4.89)***
Lag ROCE 0.62
(6.45)***
Lag turnover/employee 0.89
(14.5)***
R2 0.162 0.174 0.128 0.143 0.005 0.192 0.105 0.275 0.432 0.522 0.771
* p ≤ 0.10, ** p ≤ 0.05, *** p ≤ 0.01
t value in brackets
159
Table 6-7: Lean bundles interaction and firm performance- results from PLS-SEM
Endogenous variables
ABC Process
factor
Physical
structure
factor
Customer
value
factor
Error
prevention
factor
Customer
satisfaction
Waste
elimination
Process time
reduction
PM ROCE TE
Standardised coefficient (β)
Differentiation 0.31
(3.83)***
0.18
(1.84)* 0.03 (0.26)
0.23
(2.4)** 0.07 (0.58)
Cost leadership
-0.3 (3.09)***
0.3 (2.75)*** 0.29 (2.68)*** 0.19 (1.64) -0.04 (0.34)
ABC
0.23 (2.07)** 0.28 (2.79)***
0.19 (2)** -0.02 (0.12)
-0.01 (0.08)
-0.04 (0.43) 0.01 (0.07)
0.03 (0.34)
-0.01 (0.08)
0.09 (1.27)
Motivation factor
0.23 (2.16)**
0.25 (2.17)** 0.27 (2.6)***
0.2 (2.1)**
0.03 (0.32)
0.03 (0.58)
Human factor
0.05
(0.44)
0.01
(0.13) 0.1 (1.02)
0.04
(0.46)
0.26
(2.85)***
0.01
(0.19)
Process factor
0.01
(0.11)
0.05
(0.46) 0.06 (0.55)
0.01
(0.06)
0.14
(1.47)
-0.01
(0.13)
Physical structure factor
0.33
(2.79)***
0.16
(1.36) 0.2 (1.8)*
0.01
(0.09)
-0.11
(1.15)
-0.01
(0.09)
Customer value factor
0.05
(0.38)
-0.14
(0.88) 0.06 (0.51)
0.13
(1.24)
0.03
(0.23)
-0.05
(0.58)
Error prevention factor
0.1 (1.02)
0.01 (0.08) 0.05 (0.5)
-0.08 (0.84)
-0.01 (0.17)
0.01 (0.1)
Age
0.08 (0.83)
0.22 (1.96)** 0.17 (1.61)
-0.03 (0.31)
-0.12 (1.27)
0.08 (1.3)
Size
0.06
(0.5)
-0.03
(0.33) 0.06 (0.58)
-0.14
(1.41)
0.06
(0.71)
Lag profit margin
0.42
(4.04)***
Lag ROCE
0.59
(6.94)***
Lag turnover/employee
0.85
(10.56)***
Motivation factor * Process factor
0.25 (2.79)***
0.14 (1.26) 0.14 (1.05)
0.18 (1.71)*
0.18 (2.29)**
-0.08 (1.18)
Human factor * Process factor
-0.06 (0.56)
0.22 (2.3)** 0.18 (1.7)*
0.18 (1.71)*
0.13 (1.33)
-0.04 (0.5)
Motivation factor * Physical structure factor
0.03 (0.26)
0.23 (2.1)** 0.2 (1.83)*
0.15 (1.5)
0.14 (1.54)
0.03 (0.53)
Human factor * Physical structure factor
-0.03 0.1 -0.07 (0.64) 0.26 -0.01 0.09
160
(0.19) (0.88) (2.15)** (0.07) (1.11)
Motivation factor * Customer value factor
0.25
(2.59)***
0.19
(1.52) 0.05 (0.39)
0.14
(1.64)
0.18
(2.26)**
-0.04
(0.58)
Human factor * Customer value factor
0.12
(0.88)
0.1
(0.77) 0.11 (0.99)
0.05
(0.44)
-0.07
(0.61)
-0.04
(0.55)
Motivation factor * Error prevention factor
0.27 (2.44)**
0.27 (2.39)** 0.2 (1.69)*
-0.04 (0.32)
-0.05 (0.71)
0.09 (0.94)
Human factor * Error prevention factor
-0.06 (0.55)
0.15 (1.39) -0.01 (0.07)
0.11 (0.98)
0.03 (0.34)
-0.05 (0.62)
R2 0.162 0.174 0.128 0.143 0.005 0.44 0.375 0.459 0.593 0.591 0.789
* p ≤ 0.10, ** p ≤ 0.05, *** p ≤ 0.01
t-statistic in brackets
161
6.4.1 Lean bundles, contextual variables and firm performance
This study expects that both the social and technical bundles of lean service have an
independent positive impact on both operational and financial performance. Further, it is
expected in this research that the technical and social bundles of lean service will interact
to improve firm performance beyond that improvement achieved by each set of bundles
separately. In addition, three contextual variables (i.e. firm size, firm age and ABC) are
assumed to have a relationship with firm operational and financial performance. This
subsection presents the results of testing for the expected associations between lean
bundles, their interaction, contextual variables and firm operational and financial
performance.
6.4.1.1 The main effect of lean technical bundles on firm performance
The first two hypotheses in this research study (H1a and H1b) predict a direct positive
relation between the four technical bundles of lean service (i.e. process factor, physical
structure factor, customer value factor and error prevention factor) and firm operational
and financial performance. The results reported in Table 6-6 indicate that only two factors
are positively associated with only operational performance. Specifically, the customer
value factor is found to have a positive relationship with customer satisfaction (β = 0.27, p
< 0.05). In addition, the error prevention factor is positively related to process time
reduction (β = 0.18, p < 0.10). This provides partial support for H1a. However, none of the
four technical lean bundles is found to be related to the three financial indicators used in
this study which does not support H1b.
6.4.1.2 The main effect of lean social bundles on firm performance
As stated in H2a and H2b, the two social bundles of lean service are anticipated to have a
direct positive relation with firm operational and financial performance. Table 6-6 reveals
a direct positive association between the motivation factor and process time reduction. The
two social bundles are found to have a direct positive relationship with financial
performance. Specifically, while the motivation factor has a direct positive association
with profit margin (β = 0.20, p < 0.05), the human factor is directly and positively related
162
to ROCE (β = 0.28, p < 0.01). Collectively, these findings bear support for H2b but only
partial support for H2a.
6.4.1.3 The synergy between the social and technical bundles of lean service
The social and technical bundles of lean service are expected to collaborate in improving
firm performance over and above the improvement achieved from each set separately as
indicated by H3a and H3b. Inspecting the results reported in Table 6-7 lends support to
this notion. It is found that different social and technical bundles positively interact to
improve different dimensions of firm operational and financial performance. The results in
Table 6-7 demonstrate that while the process factor positively interacts with the
motivation factor to improve customer satisfaction (β = 0.25, p < 0.01), profit margin (β =
0.18, p < 0.10) and ROCE (β = 0.18, p < 0.05), it also positively interacts with the human
factor to improve waste elimination (β = 0.22, p < 0.05), process time reduction (β = 0.18,
p < 0.10) and profit margin (β = 0.18, p < 0.10). Further, the physical structure factor
collaborates with the motivation factor to enhance waste elimination (β = 0.23, p < 0.05)
and process time reduction (β = 0.20, p < 0.10). The physical structure factor also interacts
with the human factor to improve profit margin (β = 0.26, p < 0.05). The customer value
factor seems to be less effective as it only interacts with motivation factor to improve
customer satisfaction (β = 0.25, p < 0.01) and ROCE (β = 0.18, p < 0.05). Finally, the
error prevention factor is shown to merely interact with the motivation factor to improve
all operational performance factors, namely customer satisfaction (β = 0.27, p < 0.05),
process time (β = 0.20, p < 0.10) and waste elimination (β = 0.27, p < 0.05). These
findings collectively validate H3a and H3b stated in this research.
6.4.1.4 Firm age and performance
The effect that firm age has on firm performance has not been clear given the contrasting
theoretical and empirical evidence presented in subsection 3.5.2.1. Therefore, H4a and
H4b have been stated in the non-directional form. However, the results in Table 6-6
implies that firm age has a positive relation with operational performance represented by
waste elimination factor (β = 0.20, p < 0.10). No evidence of relationship between firm
163
age and financial performance is detected. Consequently, H4a is supported while H4b is
not supported.
6.4.1.5 ABC and performance
The use of ABC is expected to directly improve the operational and financial
performance of firms as expected in H5a and H5b. however, the results reported in both
Table 6-6 and Table 6-7 do not support neither of these hypotheses. ABC is not
significantly related with any of the operational and financial performance factors.
Therefore, H5a and H5b are rejected.
6.4.1.6 Firm size and performance
Like the effect of firm age on performance, the effect of firm size has not been clear given
the contrasting theoretical and empirical evidence presented in subsection 3.5.2.3.
Therefore, H6a and H6b have been stated in the non-directional form. However, the
results in Table 6-6 indicate a positive relation between firm size and financial
performance represented by ROCE (β = 0.13, p < 0.10). No significant association
between firm size and operational performance is detected. In consequence, H6a is
rejected and H6b is supported.
6.4.2 Lean technical bundles, ABC and business strategy
The second set of hypotheses concerns the impact of the use of ABC and business strategy
adopted on the technical bundles of lean service. Therefore, this subsection will present
the results of testing the last five hypotheses in this study which are H7-H11.
6.4.2.1 Differentiation strategy and lean technical bundles
As the literature review has revealed, it is expected that adopting the differentiation
strategy will have a direct and positive impact on the implementation of lean technical
bundles (H7). Inspection of the results reported in Table 6-6 highlights a direct positive
impact of adopting the differentiation strategy on process factor (β = 0.18, p < 0.10) and
customer value factor (β = 0.23, p < 0.05). As a result, H7 is proved by this empirical
result.
164
6.4.2.2 Cost leadership strategy and lean technical bundles
H8 predicts a direct and positive influence of adopting a cost leadership strategy on the
implementation of lean technical bundles. The results shown in Table 6-6 demonstrate a
strong positive effect of adopting the cost leadership strategy on the implementation of the
process factor (β = 0.30, p < 0.01) and physical structure factor (β = 0.29, p < 0.05). This
apparently supports H8.
6.4.2.3 Differentiation strategy and ABC
Adoption of the differentiation strategy is expected to be positively association with the
implementation of ABC as stated in H9. This hypothesis is strongly supported given the
large and significant path coefficient (β = 0.31, p < 0.01) form differentiation strategy to
ABC reported in Table 6-6.
6.4.2.4 Cost leadership strategy and ABC
It is assumed in this research that companies adopting the cost leadership will less likely to
rely on ABC as can be indicated from H10. This hypothesis is strongly supported. Table
6-6 provides evidence on the negative relation between the adoption of the cost leadership
strategy and the use of ABC (β = -0.30, p < 0.01).
6.4.2.5 ABC and the technical bundles of lean service
The last hypothesis in this study anticipates a positive association between the use of ABC
and the implementation of the technical bundles of lean service (H11). The results of both
Table 6-6 and Table 6-7 are consistent in supporting this hypothesis. The use of ABC is
positively associated with three technical factors, namely process factor (β = 0.23, p <
0.05), physical structure factor (β = 0.28, p < 0.01) and customer value factor (β = 0.19, p
< 0.10). Table 6-8 summarises the outcome of hypotheses testing. As can be seen from
Table 6-8, all but four hypotheses have been supported by the empirical analysis
conducted in this study.
165
Table 6-8: Summary table of hypotheses testing
No. Hypotheses
Expected
results
Empirical
results
H1a
There is a direct positive relationship between LTPs and
operational performance of service firms. +
Partial
support
H1b There is a direct positive relationship between LTPs and financial
performance of service firms. + Rejected
H2a
There is a direct positive relationship between LSPs and
operational performance of service firms. +
Partial
support
H2b There is a direct positive relationship between LSPs and financial
performance of service firms. + Supported
H3a There is a synergy between LTPs and LSPs in improving
operational performance of service firms. + Supported
H3b There is a synergy between LTPs and LSPs in improving
financial performance of service firms. + Supported
H4a
There is a direct relationship between firm age and operational
performance of service firms. +/-
Supported
(+)
H4b There is a direct relationship between firm age and financial
performance of service firms. +/- Rejected
H5a There is a positive relationship between the use of ABC and
operational performance of service firms. + Rejected
H5b There is a positive relationship between the use of ABC and
financial performance of service firms. + Rejected
H6a There is a direct relationship between firm size and operational
performance of service firms. +/- Rejected
H6b
There is a direct relationship between firm size and financial
performance of service firms. +/-
Supported
(+)
H7
There is a direct positive relation between differentiation strategy
and the implementation of LTPs. + Supported
H8
There is a direct positive relation between cost leadership strategy
and the implementation of LTPs. + Supported
H9
The differentiation strategy is positively related to the adoption of
ABC + Supported
H10 The cost leadership strategy is negatively related to the adoption
of ABC. - Supported
H11 There is a positive relation between the use of ABC and the use of
LTPs. + Supported
166
6.5 Summary of chapter six
Chapter six focused on developing and assessing the measurement and structural model of
this study. First, the measurement model was developed and its constructs were assessed
in terms of their reliability and validity. Second, having the measurement model passed
the assessment tests, the structural model was estimated taking into account the unique
feature of PLS-SEM of using standardised latent variable scores which is found to affect
interaction testing. Finally, all hypotheses developed in chapter have been tested and the
results were reported.
167
Chapter 7 : Discussion and implications
7.1 Introduction
In the previous chapter, the research hypotheses were tested and the results were reported.
This chapter will discuss these results in more detail which allows for answering the
research questions and achieving the aim and objectives of this research study. The
discussion of the empirical results reported in the previous chapter will proceed in two
stages. First, the set of hypotheses focusing on the relationships between lean bundles,
contextual variables and firm performance will be further explored. Second, discussions of
the role of ABC and business strategy in the implementation of the technical bundles of
lean service will follow.
7.2 Lean service bundles, contextual variables and firm performance
This section presents a detailed discussion of the results achieved in the previous chapter
in regard with the hypotheses associated with the impact of lean bundles and contextual
variables on firm performance. This implies a specific focus of this section on hypotheses
H1a-H6b as presented in Table 3-5 in the summary section of chapter 3.
7.2.1 The direct relationship between LTP factors and firm performance
The empirical results in this study partially support the direct positive impact of LTPs on
operational performance (H1a) while they do not support the same impact on financial
performance (H1b). The results indicate that the observed positive association between
LTPs and operational performance is driven by two lean technical factors, i.e., customer
value factor and error prevention factor. The customer value factor is proved effective in
improving customer satisfaction while the error prevention factor seems important for
improving process time. This finding is not surprising given the higher emphasis placed
by the sample firms on these two technical factors as concluded in subsection 5.4.1 and
Table 5-6. In addition, having two (out of four) significant LTPs factors conveys that
various lean practices play role in improving performance. Hence, service firms focusing
168
on a limited number of those improvement practices may only realise limited
advancement in performance. This conclusion accords with findings reported recently by
Kim et al. (2012) and de Leeuw and van den Berg (2011).
Furthermore, two technical factors i.e. process factor and physical structure factor are
found to have no association with any of the three operational performance factors. This
conclusion should not imply that these factors are not significant elements of lean service
and can be simply neglected. Inspecting Table 6-5 reveals that, on average, the sample
service firms are at the early stage of implementing LTPs with higher emphasis on the
customer value and error prevention factors at the account of the process and physical
structure factors. If the sample service firms are following the five principles of lean
service introduced by Womack and Jones (1996) in order (see section 2.3), they should
first implement those practices that help understand the value from customer perspective.
As Table 5-6 shows service firms have focused more on the customer value factor which
includes practices such as VSM, quality function deployment and policy deployment.
Consulting the glossary sheet in Appendix 3 for a definition of those practices can
highlight the critical role of these practices in understanding the value from the customer
point of view and linking this value to a firm strategy. In consequence, the results in this
study indicate that the sample firms are on the correct track of their LTPs implementation
by focusing first on customer value identification which then will be followed by
improving processes and firm structure in light of this value (Shah and Ward, 2007, 2003;
Womack and Jones, 1996). Despite that, the lack of association between the two technical
factors and operational performance concurs with previous research. For example,
Sakakibara et al. (1997) and Bonavia and Marin (2006) conclude that LTPs do not have
any direct association with improvement at the operational level.
Further, it has been realised that none of the four technical factors of lean service has a
direct positive impact on financial performance as expected in H1b. This finding goes
along with the finding of the few previous similar studies (e.g. Fullerton and Wempe,
2009; González-Benito, 2005; Patterson et al., 2004). Fullerton and Wempe (2009)
examine the direct positive impact of three technical practices on return on sales using
169
data from 121 manufacturing firms and have not detected the anticipated positive impact.
In a similar vein, González-Benito (2005) investigates the direct impact of technical
practices on return on assets using data from 186 manufacturing firms. However, the
results of González-Benito’s (2005) study demonstrate that the technical practices alone
are not capable of improving profitability. Similar findings have been documented by
Patterson et al. (2004) who also have not been able to capture the direct impact of lean
technical practices on the of sales per employee and profit.
In contrast to the above findings demonstrating a non-existence of direct influence of
technical practices on financial performance, Kaynak (2003) reports empirical evidence
supporting the capability of LTPs in directly improving the operational and financial
performance of both manufacturing and service firms. Similarly, Alsamdi et al. (2012)
reveal a direct positive association between the technical side of lean service and the
operational and financial performance of adopters. Agarwal et al. (2013) use data from
152 manufacturing firms to test the assumed positive impact of lean practices on both
operational and financial indicators. The results prove that while the lean index has a
direct positive influence on sales, profit and profit margin, it has no effect on return on
equity and negative effect on sales growth. However, it should be noted that the
documented positive impact of LTPs on financial performance in the first two studies has
been obtained based on subjective data from respondents rather than secondary data like
the one used in this research.
The lack of association between the technical factors and financial performance can also
be attributed to the time lag needed for the effect of an improvement practice to be
reflected in financial statements (Shafer and Moeller, 2012; Birdi et al., 2008; Mohrman et
al., 1995). As the sample service firms in this study are found to be at early stage of their
LTPs implementation, it is possible that the effect of these practices has not accumulated
to a level that can be seen in the financial statements of those firms. Some researchers
have suggested a time lag of three to six years for the effect of an improvement practice to
materialise (Shafer and Moeller, 2012; Birdi et al., 2008; Powell, 1995). As a result, future
170
research is encouraged to adopt a longitudinal research methodology to supplement the
current literature with more insights on this aspect.
To conclude, collectively, the results of testing H1a and H1b can imply that at least some
of the technical practices of lean service are capable of generating benefits to service firms
that outweigh the cost of their implementation.
7.2.2 The direct relationship between LSP factors and firm performance
In line with the theoretical argument leading to H2a and H2b, the empirical results of this
study support a direct positive association between operational and financial performance
factors on the one hand, and LSP factors on the other hand. Specifically, the motivation
factor of the social side of lean service has a strong positive relation with process time
reduction and profit margin. The human factor is found to have a direct positive
association with ROCE.
Unlike the influence of LTPs, the influence of LSPs goes beyond improving operational
performance to enhance financial performance, namely profit margin and ROCE as the
results of Table 6-6 demonstrate. This confirms the findings of previous research which
highlights the superiority of the social practices over the technical practices (e.g. Talib et
al., 2013; Samson and Terziovski, 1999; Sakakibara et al., 1997; Powell, 1995). While
Dabhilkar and Åhlström (2013) and Pont et al. (2008) have not been able to capture the
direct positive influence of LSPs on the performance of adopters, the majority of
researchers have disagreed with such findings. For example, Cua et al. (2001) employ data
from 163 manufacturing firms and report evidence implying that the social practices are
positively related to operational performance. Shah and Ward (2003) conduct a survey
study of 1757 manufacturing plants to study the effect of lean system represented by four
bundles, namely JIT, TQM, TPM and HRM, on the operational performance. Their
findings confirm the direct positive influence of each bundle on operational performance
of adopters. Birdi et al. (2008) conclude that while LTPs can be transplanted easily, the
knowledge created by LSPs to companies cannot and so their effect will be superior to the
impact of LTPs.
171
7.2.3 The synergistic effect of LSPs and LTPs on firm performance
The synergy expected between the technical and social factors of lean service has been
detected in the data of this research study; H3a and H3b have been accepted. This implies
that each side of lean service does play an indirect role in improving firm performance
through enhancing the performance outcome of the other side. Consequently, the
theoretical argument developed in sub-section 3.5.3.1 in connection with the validity of
the STS mechanism has proved true in this study.
Although none of the technical factors has a direct positive relation with financial
performance (subsection 7.2.1), three technical factors (i.e. process factor, physical
structure factor and customer value factor) interact with the two social factors to improve
profit margin and ROCE. In addition, the process factor and physical structure factor
which have not influenced the operational performance alone also interact with the two
social factors to enhance all operational performance factors (i.e. customer satisfaction,
waste elimination and process time reduction). These findings emphasise the importance
of implementing the two sides of lean service together. This is so because the
simultaneous implementation of both sides will result in an improvement in both
operational and financial performance greater than the sum of their independent
improvements.
The results of this study in term of the synergy hypotheses accord with the findings of De
Menezes et al. (2010), Das and Jayaram (2007), González-Benito (2005), and Flynn et al.
(1995). Using data collected on 42 US plants, Flynn et al. (1995) forward evidence which
proves a positive interaction between the social and technical factors in improving tow
operational indicators, namely cycle time and quality performance. González-Benito
(2005) also confirms, using data from 186 manufacturing firms, the positive interaction
between social and technical factors in improving financial performance represented by
return on assets. More recently, Das and Jayaram (2007) adopt the socio-technical
perspective to examine the synergy between four lean technical practices (i.e. kanban,
group technology, JIT supply, TPM) and three HRM practices (i.e. cross-trained
172
employees, operator teams, decentralised decision-making). Based on data from 322
manufacturing firms, the authors demonstrate the expected synergy between the two sets
of practices on operational performance.
In their current form, the results of testing H1 (a, b), H2 (a, b) and H3 (a, b) have
significant implications for service managements. Service managers can rely on practices
from each side of lean service (LTPs and LSPs) to improve operational and financial
performance. However, the best utilisation of their resources can be achieved by investing
in both sets of practices simultaneously. That is, if a service firm has limited resources to
implement a few practices from lean service, the results suggest that the firm chooses
practices from LSPs and LTPs that are likely to collaborate and yield higher performance
improvement. For example, Table 6-7 indicates that unlike the human factor, the
motivation factor interacts with all technical factors to improve firm performance.
Therefore, with limited resource, investing in this factor along with one or more of the
technical factors seems more promising than investing first in the human factor of the
social side.
7.2.4 Firm age and firm performance
H4a and H4b have expected a direct relationship between firm age and the performance of
service firms. The results of Tables 6-6 provide support for a positive association between
firm age and the waste elimination factor. This finding validates the argument that older
firms have better knowledge and experience which enable them to run their operations
more efficiently than less experienced firms (Coad et al., 2013; Lundvall and Battese,
2000; Glancey, 1998). However, it is not necessary that such companies will enjoy higher
financial performance as firm age has not had a positive relation with financial
performance which supports the findings of previous studies (e.g. Wagner et al., 2012).
7.2.5 ABC and firm performance
H5a and H5b have anticipated that companies who adopted ABC have a better operational
and financial performance than those who adopted TAS. However, neither of these
hypotheses has been supported. This indicates that neither of these two accounting
173
systems is superior to the other in terms of improving firm performance. Similar
conclusion has been reached by some researchers (e.g. Sheu and Pan, 2009; Cagwin and
Bouwman, 2002; Mishra and Vaysman, 2001). Sheu and Pan (2009) and Mishra and
Vaysman (2001) report that both ABC and TAS can be effective but under different
circumstance. ABC will be more effective under high uncertainty levels while TAS will
be more effective under low uncertainty levels. Consequently, if firms adopt the
appropriate system based on their conditions, no difference in their performance may be
detected.
Another possibility can be the recent argument by Banker et al. (2008), who suggest that
ABC will not have a direct effect on firm performance. Rather it will improve other
capabilities such as the implementation of lean practices which in turn improve
performance. Using data from a large sample of manufacturing firms, Banker et al. (2008)
have empirically confirmed this possibility. It seems that the results of this research study
lend itself to this perspective. As has been demonstrated in subsection 6.4.2.5, adopting
ABC is found to support the implementation of LTPs which in turn are proved to improve
at least the operational performance (H1a has been supported).
7.2.6 Firm size and firm performance
Although firm size, represented by the number of employees, is proved to have no relation
with operational performance, its positive association with financial performance (i.e.
ROCE) is evident. This finding implies that although larger firms do not necessarily enjoy
better operational performance, they do at financial level. The lack of association between
firm size and operational performance found in this study concurs with earlier evidence
reported by González-Benito (2005) who also has not been able to capture such
association using data from 186 firms on four operational indicators, namely cost, quality,
flexibility and reliability. Moreover, the positive association between firm size and
financial performance confirms the notion that larger firms are likely to have more
financial resources than small firms (Jayaram et al., 2010). It also goes in line with the
findings of Wagner et al. (2012) who have used empirical data from 259 manufacturing
firms and proved a positive influence of firm size on financial performance.
174
7.3 Lean service, ABC and business strategy
This section presents a detailed discussion of the results achieved in the previous chapter
in regard to the hypotheses associated with the role of ABC and business strategy in the
implementation of LTPs. This implies a specific focus of this section on hypotheses H6-
H11 as presented in Table 3-5 in the summary section of chapter 3.
7.3.1 LTPs and Business strategy
The cost leadership and differentiation strategies have been anticipated to have a direct
positive impact on the implementation of LTPs (H7 and H8). Both hypotheses have been
supported as presented in subsections 6.4.2.3 and 6.4.2.4. However, firms adopting the
differentiation strategy are found to focus on the customer value factor and process factor
(see Table 6-6). On the other hand, firms adopting the cost leadership strategy have higher
emphasis on the process factor and physical structure factor. The difference in focus
between these two strategies provides empirical validation to Porter’s (1980) notion that
differentiators and cost leaders have different priorities.
Differentiators compete through product/service innovation and customisation where
understanding the value from customer perspective is critical (Kumar and Telang, 2011;
Kennedy and Widener, 2008; Gosselin, 1997). As a result, they have strongly
implemented the customer value factor which includes practices important for
understanding the value from customer point of view. In contrast, cost leaders compete on
the price basis and emphasise process efficiency and cost reduction to enable them
compete on the price basis (Ward et al., 2007; Frey and Gordon, 1999; Gosselin, 1997).
Therefore, they have devoted their effort to implement the process factor and physical
structure factor which include practices focusing on waste elimination at both process and
structure levels. Interestingly, by implementing the process factor, differentiators seem to
keep an eye on the efficiency of their processes even though they can usually ask for a
price premium for their innovative and customised products/services (Qi et al., 2011).
175
Recognising the emphasis on different sets of lean practices by each strategy can offer one
explanation to the mixed results found in the few empirical studies which have examined
the strategy-lean association (e.g. Qi et al., 2011; Ward et al.; 2007; Baines and Langfield-
Smith, 2003; Chenhall and Langfield-Smith, 1998). The findings propose that detecting a
significant relation between each type of strategy and lean practices depends on the lean
practices included in the study. For instance, representing lean service by practices
included in the customer value factor will support the fit between differentiation strategy
and lean found by Baines and Langfield-Smith (2003) and Chenhall and Langfield-Smith
(1998). If lean service is represented by practices contained in the physical structure
factor, the finding is expected to advocate the match between cost leadership strategy and
lean found by Qi et al. (2011). The preceding finding implies that researchers should avoid
a restricted representation of lean service by a few practices when studying the strategy-
lean association.
7.3.2 Business strategy and ABC
H9 and H10 state a positive association between the differentiation strategy and ABC and
a negative association between the cost leadership strategy and ABC, respectively. The
results of Table 6-6 have confirmed both hypotheses. These results empirically confirm
the notion which indicates that given the differences in priorities and environments in
which differentiators and cost leaders operate, differentiators will more likely prefer ABC
which generates more accurate and detailed financial and non-financial information and
helps in understanding value adding activities necessary to enhance products/services
differentiation (Khataie et al., 2011; Cagwin and Bouwman, 2002; Chenhall and
Langfield-Smith, 1998). Similar conclusion has been reported in this stream of literature.
For instance, Baines and Langfield-Smith (2003) use data from 140 manufacturing firms
and forward evidence suggesting that adopting differentiation strategy will lead to
increased use of advanced MASs including ABC. In a similar vein, Gosselin (1997) uses
data from 161 manufacturing firms and find the adoption of differentiation strategy to be
associated with implementation of ABC. In addition, Frey and Gordon (1999) and
Chenhall and Langfield-Smith (1998) demonstrate through empirical evidence that ABC
benefits differentiators more that cost leaders.
176
7.3.3 LTPs and ABC
As has been expected by H11, the adoption of ABC has been association with
implementation of LTPs. This result has validated the significant direct role of ABC in
influencing the implementation level of LTPs. More specifically, This advocates the
notion that as ABC improves the visibility of what truly drives cost and consumes
resources at activity level, it helps expose areas for improvements by identifying non-
value adding activities which encourages the implementation of lean service practices to
eliminate them (Khataie and Bulgak, 2013; Chiarini, 2012; Li et al., 2012; Maiga and
Jacobs, 2008; Clarke and Mullins, 2001). In addition, when taken together the positive
ABC-lean association (H11) and lean-performance relation (H1a) found in this study bring
to light a new perspective into the ABC-performance literature (e.g. Sheu and Pan, 2009;
Cagwin and Bouwman, 2002; Kennedy and Affleck-Graves, 2001; Mishra and Vaysman,
2001). This new perspective suggests that ABC is capable of contributing to firm
performance indirectly through helping companies develop other organizational
capabilities (e.g. lean service) which, in turn, can improve performance. In consequence,
these findings imply that by implementing ABC, firms can kill two birds with one stone.
First, they enjoy better understanding of how and where their resources are consumed
which is important for informed decision making. Second, they become in a better
position to improve their processes and performance through implementing lean service
practices.
The important role of ABC in the lean context found in the current study emphasises
earlier evidence in the literature. For instance, Al-Omiri and Drury (2007) in a survey
study of 176 UK manufacturing and service firms highlight a positive relationship
between the implementation of lean system and the level of cost system sophistication. In
addition, Innes and Mitchell (1995) survey the largest UK companies and find them to
depend on the measures generated from ABC to support other improvement initiatives
such as continuous improvement, TQM and JIT. Adam (1996) argues that when ABC is
used in conjunction with other process improvement systems such as quality and lean
system, companies enjoy a higher level of benefits. Khataie and Bulgak (2013) use a
177
system dynamics modelling tool and reveal the importance of ABC in achieving the aim
of lean system. Moreover, Banker et al. (2008) report empirical evidence indicating that
companies adopting ABC are more likely to adopt lean practices in their operations.
7.3.4 ABC, business strategy and LTPs
When taken together the results of testing H7-H11 offer another significant insight into the
relationship among ABC, business strategy and LTPs. In addition to the direct relation
between business strategy and LTPs (H7 and H8), there is also an indirect relation acting
through the accounting variable (ABC) as H9, H10 and H11 have been supported. The
support found for H9 indicates that adopting differentiation strategy is positively
associated with implementation of ABC which overcomes TAS in uncertain environments
and when more customized products/services are produced (Khataie et al., 2011; Sheu and
Pan, 2009; Gosselin, 1997; Cooper and Kaplan, 1992). Being supported, H9 and H11
highlight the indirect positive association between differentiation strategy and lean service
through the intervening variable ABC. However, given that H7 (direct relation between
differentiation and lean service) is supported; the variable ABC only partially mediates the
differentiation-lean association.
In contrast to differentiators, cost leaders are found to rely more on TAS given the support
found for H10 which proposes a negative relation between cost leadership strategy and the
dummy variable ABC constructed as 1 for ABC and 0 for TAS. Relying on TAS by cost
leaders is not surprising given their tendency to avoid investing in innovations and
producing a relatively limited range of standardized products/services (Chenhall, 2003;
Lamminmaki and Drury, 2001; Gosselin, 1997). However, being supported, H10 and H11
lead to the conclusion that cost leadership strategy has an indirect negative relation with
lean service which contradicts its direct positive relation (H8). In other words, the
accounting variable (ABC) in the model suppresses the relationship between cost
leadership strategy and lean service leading to a case of inconsistent mediation (Taylor et
al., 2008; MacKinnon et al., 2000). Inconsistent mediation occurs when the direct and
indirect effects have opposite signs (Taylor et al., 2008; MacKinnon et al., 2000).
Consequently, the total effect of cost leadership strategy on lean service will depend on
178
the magnitude of each effect (i.e. direct and indirect). When the magnitudes of the direct
and indirect effects are close to each other or identical, the total effect will be zero or close
to zero. This can be one reason behind the lack of association between the cost leadership
strategy and lean practices found by studies which did not account for the role of ABC in
that relation. Therefore, the implication of this finding is that researchers should include
the effect of the accounting system used by companies to better understand the
relationship between business strategy and lean practices. In addition, while TAS may
provide less distorted cost information for cost leaders operating in a stable environment
and producing a relatively limited set of standardized products, it does not help in
developing other organizational capabilities (i.e. lean service) which have a direct positive
impact on performance (Li et al., 2012). The two case studies reported by Cooper and
Maskell (2008) and Datar et al. (1991) clearly demonstrate the detrimental effect of TAS
on the implementation of lean practices which supports the above conclusion.
7.4. Summary of chapter seven
This chapter presented a detailed discussion of the hypotheses testing results reported in
chapter 6. In discussing the results of this research study, there has been an attempt to
position the results achieved for each of the research hypotheses within the relevant extant
literature so that differences have been highlighted and implications have been deduced.
The findings of this research indicated that at least some LTPs can improve operational
performance of service companies. LSPs were found to have a direct positive relationship
with both operational and financial performance of service firms. More importantly, there
was a positive interaction between LTPs and LSPs which improved firm performance over
and above the improvement that could be achieved from each separately. Moreover,
activity-based costing system was revealed to be critical in overcoming the limitations of
the traditional accounting system (TAS) and consequently encouraging the
implementation of lean technical practices. This was a significant finding as it highlighted
the role of ABC in improving firm performance indirectly through other organisational
capabilities (i.e. lean practices) which, in turn, influenced performance. Finally, In respect
of business strategy, both differentiation and cost leadership strategies had a direct
179
positive association with lean service. However, while the differentiation strategy also had
an indirect positive association with lean service through ABC, the cost leadership
strategy had an indirect negative relationship with lean service due to its negative relation
with ABC.
180
Chapter 8 : Conclusions
8.1 Introduction
This study aims to uncover the full potential of lean service in improving firm
performance and the role of two contextual variables (i.e. ABC, business strategy) in the
adoption of lean service. To address this aim, the current study makes use of two well
known theories (i.e. STS and CT) to develop a theoretical model and a set of research
hypotheses. The model views lean service as a socio-technical system with two sides:
technical side and social side. The mechanism of the STS is then relied on to expand the
traditional lean-performance model focusing mainly on the direct and additive effect of
lean bundles by highlighting the possible synergy between the two sides of lean service.
By doing so, this study probes deeper in the mechanism through which lean service is
expected to influence firm performance. Consequently, the model assumes each side of
lean service to play two roles in influencing firm performance: a direct role and an indirect
role through enhancing the impact of the other side.
Moreover, adopting the CT sets the foundation for underlining the potential effect of two
contextual variables (i.e. ABC, business strategy) on the adoption level of the technical
side of lean service usually neglected in the current literature. In addition, the model
developed in this study extends the focus of the few studies which examined only the
additive impact of ABC and business strategy on lean practices. By highlighting the
relationships among these three variables, the model emphasises the need to examine the
possibility of the indirect effect, acting through the other variables, of each variable on
lean practices along with their direct effect.
To test the theoretical model a positivist approach is adopted, and a cross-sectional survey
methodology is employed in the UK service context. The UK is an appropriate context for
the current study given the established service nature of its economy. The data are
collected using multiple methods including questionnaire instrument and archive database.
181
The most appropriate statistical techniques are adopted for data analysis and hypotheses
testing including factor analysis and PLS-SEM.
The second section of this chapter will summarise the research findings of the data
analysis and the hypotheses testing process. In the third section, the research questions
will be revisited in an attempt to provide answers to these questions given the findings of
this research study. The main conclusions of this study will be provided in the fourth
section. The fifth section is devoted to highlight the main contributions of this study.
Finally, the last section of this chapter will report the limitations of this study and provide
directions for future research.
8.2 Summary of research findings
Testing the direct and synergistic impact of the two sides of lean service on firm
performance while controlling for the effect of ABC, industry, past performance, firm size
and firm age indicates the following:
i. LTPs are found to have a direct positive association with operational performance
but not with financial performance indicators.
ii. Adopting LSPs are proved to have a direct positive relationship with operational
and financial performance indicators. This relationship is stronger than that
between LTPs and firm performance.
iii. The effect of the expected collaboration (synergy) between the two sides of lean
service is empirically supported by the data collected in this study.
iv. In terms of the impact of the contextual variables on firm performance, the use of
advanced MAS (i.e. ABC) is proved to have no direct impact on operational and
financial performance. Firm size is found to have a positive relation with financial
performance but not with operational performance. Firm age is shown to have a
positive influence on firm operational performance rather than financial
performance.
182
In relation to the impact of ABC and business strategy on the technical side of lean
service, the results support the following conclusions:
i. The use of ABC has a direct positive relation with the level of LTPs adoption.
ii. The cost leadership strategy is shown to have a direct positive association with
LTPs. However, its negative indirect association with LTPs acting through ABC
(intervening variable) has reduced the magnitude of its direct positive relation with
LTPs.
iii. The differentiation strategy is found to have positive direct and indirect effect
acting through ABC on the implementation of LTPs.
As has been reported above, a number of relationships as proposed by the theoretical
model are supported in this study. Moreover, the results also provide support to the
argument highlighting the need to focus on not only the operational performance but also
the financial performance given that some variables (e.g. LSPs) in the model have a
relationship with both. In addition, although the implementation of ABC is found to have
no direct relation with firm performance, its indirect association with performance is
evident through its effect on lean practices which are, in turn, related to firm performance.
In light of the research findings reported above, the research questions can now be
revisited in an attempt to answer those questions as presented in the next section.
8.3 Revisiting the research questions
This research study sought to answer mainly two research questions as stated in the first
chapter. These two questions are:
1. Do lean practices have an additive (direct) and/or non-additive (indirect) impact on
operational and/or financial performance of service firms?
2. Does each of the following contextual variables: ABC and business strategy affect
lean service practices? And if so, is the effect direct, indirect or both?
8.3.1 The first research question
The first research question constructively questions the traditional lean-performance
model focusing mainly on the additive impact of lean practices on mainly firm operational
183
performance. The model of the current study suggests expanding the traditional model by
empirically investigating the non-additive (synergy) impact of lean practices on both
operational and financial performance as only very limited information is available to date
in this respect.
To answer the first research question, lean service is viewed as a socio-technical system
where each of its two sides is expected to directly influence firm performance and
indirectly through the expected collaboration between them. PLS-SEM analysis is used to
test the corresponding hypotheses. The results suggest a direct positive impact of the two
sides of lean service on firm operational performance, although a stronger impact of the
social side is evident. In addition, while the effect of the social side extends to the
financial performance, the impact of the technical side does not. More importantly, the
non-additive effect (synergy) of each side of lean service is statistically supported in this
study. Therefore, the simultaneous implementation of the technical and social sides of lean
service is expected to result in an improvement in firm performance greater than the sum
of the independent improvements attained from each side separately.
In the current form, the findings of this study support the presence of a direct and additive
impact of the two sides of lean service on both operational and financial performance as
well as synergistic impact resulting from the interaction between the two sides.
8.3.2 The second research question
The second research question addresses the nature of the likely impact of ABC and
business strategy on the technical side of lean service. The results suggest a direct positive
impact of ABC and business strategy (differentiation and cost leadership) on the
implementation of LTPs. However, while the differentiation strategy also had a positive
indirect impact on LTPs acting through ABC, the cost leadership strategy is found to have
a negative indirect effect on LTPs acting through ABC. In other words, while ABC is
found to partially mediate the differentiation-LTPs association, it suppresses the cost
leadership-LTPs association leading to a case of inconsistent mediation.
184
In conclusion and as a response to the second research question, ABC has a direct positive
impact on LTPs, while cost leadership strategy and differentiation strategy have both
direct and indirect influence on LTPs acting through ABC.
8.4 Research Contributions
This study aims at probing the mechanism of lean service in improving firm performance
and investigating the role of ABC and business strategy in the adoption of lean service. To
achieve this aim and associated objectives:
i. A theoretical model has been developed by integrating two well known theories,
namely STS and CT,
ii. The model has been empirically examined using data collected from UK service
companies.
Consequently, the contributions of this study can be realised at different levels, namely
theoretical, methodological and empirical levels.
At the theoretical level, this study develops a conceptual framework which crosses
different streams of literatures mainly, lean system literature, management accounting
literature with focus on ABC, business strategy literature and human resources
management literature. Unlike previous studies, by integrating the perspective of STS and
CT, the model (i) highlights not only the direct effect of each of the lean service sides on
firm performance but also the potential synergy between the two sides, (ii) brings to light
the direct impact of ABC and business strategy on LTPs and the intervening role of ABC
due to which the business strategy is assumed to have also an indirect influence on LTPs,
and (iii) offers an alternative view on how ABC can improve firm performance by
enhancing other organisational capabilities which are expected to improve performance .
At the methodological level, unlike previous studies, this study includes a large number of
lean service practices and contextual variables to report more precisely on the lean-
performance association. In addition, the inclusion of the financial performance dimension
-measured by secondary data- in the model in addition to the operational performance is
critical to understand the full capability of lean service in improving firm performance.
185
Finally, employing powerful statistical techniques like factor analysis and PLS-SEM
provides more credibility to the results reported in this study.
At the empirical level, this study is conducted in the UK service sector. As such, this study
is one of the very few studies that have reported on lean service and examined how the
adoption of ABC and specific business strategy can affect its implementation using
empirical survey data from another context than manufacturing.
Finally, the findings of the current research have significant implications for both practice
and academia. First, the interaction between the two sides of lean service detected in this
study should encourage service firms to avoid focusing on the practices of only one side.
A better outcome can be expected from implementing practices from both sides. Second,
service firms should also be aware of the effect of their accounting system on the
implementation of lean service practices. In contrast to the TAS, ABC helps service firms
distinguish between value adding and non-value adding activities which may motivate and
justify the need for lean service practices to eliminate the waste associated with the non-
value adding activities. In addition, researchers are encouraged to take into consideration
the role of the accounting system when examining the strategy-lean association. By doing
so the current study has offered a better understanding of the lack of relationship between
the cost leadership strategy and lean practices found in the existing literature.
8.5 Limitations and Directions for Future Research
Like all other research studies, a number of limitations of this study can be highlighted
which can be addressed in future research. These limitations can be classified into three
groups, namely theoretical, methodological and empirical limitations.
At the theoretical level, given the complexity of the model developed in this study, only
the synergy between the two sides of lean service is conceptualised and empirically tested.
Therefore, no attempt has been made to theorise the possible interaction between any of
the contextual variables and both or either of the two sides of lean service. Similarly, only
two contextual variables (i.e. ABC and business strategy) are examined in terms of their
186
effect on LTPs. Consequently, other contextual variables such as unionisation, service
process type and firm age can be the focus of future research to investigate their impact on
lean service.
At the methodological level, the cross-sectional nature of this research prevents definitive
statements about causality between dependent variables (DVs) and independent variables
(IVs) involved in this research. Future research with longitudinal nature would advance
the findings of this study and provides unique information on the sustainability of the
impact of lean service on firm performance over time. Further, given the size of the
sample obtained in this research study, obtaining a larger sample would allow for more
robust results to be obtained. In addition, another limitation arises from using subjective
measures and single informant to collect data on both DVs and IVs. This method of
measurement is argued to introduce the possibility of higher measurement error and
inflated association between criterion and predictor variables. Despite that, subjective
measures whether single-item or multiple-item have been widely used in operations
management literature (Fullerton et al., 2013; Shah and Ward, 2007, 2003; Fullerton et al.,
2003; Cua et al., 2001).
At the empirical level, this study is limited to medium and large companies only.
Therefore, the results may not be generalisable to small companies. In addition, the study
is also limited to the UK service sector. Consequently, future research may replicate this
study in other service contexts than the UK. It should also be noted that several practices
of lean service were dropped as they had not been reported by at least five studies. Future
research may focus on these practices to examine their effectiveness and relation with
other contextual variables.
Finally, as shown in Table (4-6), a relatively large number of service firms (22% of non-
respondents) indicated irrelevancy of the questionnaire to their firms. This result points to
the high reluctance of service managers to experiment with lean service practices despite
the increasing level of literature encouraging them to do so. Therefore, valuable future
work can improve our knowledge on reasons behind this reluctance of service managers to
187
adopt lean service practices. Further, it has been out of the scope of this study to examine
whether there is a specific sequence in the implementation of LTPs and LSPs that leads to
the improved performance. Little work has been done in the literature that tries to establish
best models for best outcome of lean system. Future attempts focusing on this point can
allow for critical theoretical and practical implications to be realised.
188
References
Abdelaziz, E.A., Saidur, R. and Mekhilef, S. (2011). A review on energy saving strategies
in industrial sector. Renewable and Sustainable Energy Reviews, 15 (1), 150–168
Abdel-Kader, M. and Luther, R. (2008). The impact of firm characteristics on management
accounting practices: A UK-based empirical analysis. The British Accounting Review,
40 (1), 2-27
Abdi, F., Shavarini, S. K. and Hoseini, S. M. S. (2006). Glean Lean: How to use lean
approach in service industries? Journal of Services Research, 6 (special issue), 191-206
Abu Mansor, N.N., Tayles, M. and Pike, R. (2012). Information Usefulness and Usage in
Business Decision-Making: An Activity-Based Costing (ABC) Perspective.
International Journal of Management, 29 (1), 19-32
Adams, M. (1996). Activity-based costing (ABC) and the life insurance industry. The
Service Industries Journal, 16 (4), 511-526
Agarwal, R., Green, R., Brown, P.J., Tan, H. and Randhawa, K. (2013). Determinants of
quality management practices: An empirical study of New Zealand manufacturing
firms. International Journal of Production Economics, 142 (1), 130–145
Ahlstrom, P. (1998). Sequences in the implementation of lean production. European
Management Journal, 16 (3), 327-334
Åhlström, P. (2004). Lean service operations: Translating lean production principles to
service operations. Int. J. Services Technology and Management, 5 (5-6), 545-564
Ahluwalia, R. S., Srinivasan, A. and Irani, S. A. (2004). Evidence of lean. Industrial
Engineer, 36 (12), 35-40
Ahmad, S., Schroeder, R.G. and Sinha, K.K. (2003). The role of infrastructure practices in
189
the effectiveness of JIT practices implications for plant competitiveness. Journal of
Engineering and Technology Management, 20, 161-191
Alagaraja, M. (2010). Lean Thinking as applied to the adult education environment. Int. J.
Human Resources Development and Management, 10 (1), 51-62
Allway, M. and Corbett, S. (2002). Shifting to lean service: Stealing a page from
manufacturers' playbooks. Journal of Organizational Excellence, 21 (2), 45-54
Al-Omiri, M. and Drury, C. (2007). A survey of factors influencing the choice of product
costing systems in UK organizations. Management Accounting Research, 18 (4), 399-
424
Alsmadi, M., Almani, A. and Jerisat, R. (2012). A comparative analysis of Lean practices
and performance in the UK manufacturing and service sector firms. Total Quality
Management and Business Excellence, 23 (3-4), 381-396
Amoako-Gyampah, K. and Acquaah, M. (2008). Manufacturing strategy, competitive
strategy and firm performance: An empirical study in a developing economy
environment. International Journal of Production Economics, 111 (2), 575–592
Anania, L. (1987). Services: Low Productivity High Cost. Review of Business, 9 (1), 21-23
Anderson, S., and Widener, S. (2007). Doing quantitative field research in management
accounting. In C. Chapman, A. Hopwood, and M. D. Shields, Handbook of
Management Accounting Research, 1. (pp. 319-342). Oxford: Elseiver.
Appelbaum, S.H. (1997). Socio-technical systems theory: an intervention strategy for
organizational development. Management Decision, 35 (6), 452–463
Apte, U.M. and Goh, CH. (2004). Applying lean manufacturing principles to information
intensive services. Int. J. Services Technology and Management, 5 (5-6), 1-20
Arbos, L. C. (2002). Design of a rapid response and high efficiency service by lean
production principles: Methodology and evaluation of variability of performance.
190
International Journal of Production Economics, 80 (2), 169-183
Arlbjørn, J. S., and Freytag, P. V. (2013). Evidence of lean: a review of international peer-
reviewed journal articles. European Business Review, 25 (2), 174-205
Arlbjørn, J. S., Freytag, P. V. and Haas, H. (2011). Service supply chain management: A
survey of lean application in the municipal sector. International Journal of Physical
Distribution and Logistics Management, 41 (3), 277-295
Armstrong, S. and Overton, T. (1977). Estimating Nonresponse Bias in Mail Surveys.
Journal of Marketing Research, 14 (3), 396-402
Askarany, D., Yazdifar, H. and Askary, S. (2010). Supply chain management, activity-
based costing and organisational factors. International Journal of Production
Economics, 127 (2), 238–248
Atkinson, P. (2004). Creating and Implementing Lean Strategies. Management Services,
48 (2), 18-21
Auzair, S.M. and Langfield-Smith, K. (2005). The effect of service process type, business
strategy and life cycle stage on bureaucratic MCS in service organizations. Management
Accounting Research, 16 (4), 399–421
Baggaley, B. (2006). Using strategic perofrmance measurements to accelerate lean
performance. Cost Management, 20 (1), 36-44
Baines, A. and Langfield-Smith, K. (2003). Antecedents to management accounting
change: a structural equation approach. Accounting, Organizations and Society, 28 (7-
8), 675–698
Ball, D.R. and Maleyeff, J. (2003). Lean management of environmental consulting.
Journal of Management in Engineering, 19 (1), 17-24
Banker, R.D., Bardhan, I.R., & Chen, T.Y. (2008). The role of manufacturing practices in
mediating the impact of activity-based costing on plant performance. Accounting,
191
Organizations and Society, 33 (1), 1-19
Baxter, G. and Sommerville, I. (2011). Socio-technical systems: From design methods to
systems engineering. Interacting with Computers, 23 (1), 4-17
Bayo-Moriones, A., Bello-Pintado, A. and Merino-Dı´az-de-Cerio, J. (2008). The role of
organizational context and infrastructure practices in JIT implementation. International
Journal of Operations and Production Management, 28 (11), 1042-1066
Ben-Tovim, D. I., Bassham, J. I., Bolch, D., Martin, M. A., Dougherty, M. and Szwarc, M.
(2007). Lean thinking across a hospital: redesigning care at the Flinders Medical Centre.
Australian Health Review, 31 (1), 10-15
Bhasin, S. (2008). Lean and performance measurement. Journal of Manufacturing
Technology Management, 19 (5), 670-684
Bhasin, S. (2012). Performance of Lean in large organisations. Journal of Manufacturing
Systems, 31 (3), 349– 357
Bhasin, S. and Burcher, B. (2006). Lean viewed as a philosophy. Journal of Manufacturing
Technology Management, 17 (1), 56-72
Bhatia, N. and Drew, J. (2007). Applying Lean production to the Public Sector. The
McKinsey Quarterly, June, 1-5
Birdi, K., Clegg, C., Patterson, M., Robinson, A., Stride, C.B., Wall, T.D. and Wood, S.J.
(2008). The impact of human resource and operational management practices on
company productivity: A longitudinal study. Personnel Psychology, 61 (3), 467-501
Bodek, N. (2007). The ninth waste—Saying no. Available at:
http://www.industryweek.com/articles/the_ninth_waste_--_saying_no_15106.aspx
Bonavia, T. and Marin, J.A. (2006). An empirical study of lean production in the ceramic
tile industry in Spain. International Journal of Operations and Production
192
Management, 26 (5), 505-531
Bonavia, T. and Marin-Garcia, J.A. (2011). Integrating human resource management into
lean production and their impact on organizational performance. International Journal
of Manpower, 32 (8), 923-938
Bortolotti, T. and Romano, P. (2010). Enabling lean management adoption in the pure
service sector: when to implement automation and information systems?. In:
Proceedings of the 17th International Annual EurOMA Conference, Portugal, June,
2010), 1-10
Bortolotti, T. and Romano, P. (2012). Lean first, then automate: a framework for process
improvement in pure service companies. A case study. Production Planning and
Control: The Management of Operations, 23 (7), 513-522
Bowen, D. E. and Youngdahl, W. E. (1998). Lean” service in defense of a production-line
approach. International Journal of Service Industry Management, 9 (3), 207-225
Brewton, J. (2009). The lean office: Develop lean administrative procedures. Cost
management, 23 (2), 40-47
Brignall, S. (1997). A contingent rationale for cost system design in services. Management
Accounting Research, 8 (3), 325-346
Brignall, T.J. Fitzgerald, L. Johnston, R. and Silvestro, R. (1991). Product costing in
service organizations. Management Accounting Research, 2 (4), 227-248
Bruggeman, W. and Stede, W.V. (1993). Fitting Management Control Systems to
Competitive Advantage. British Journal of Management, 4 (3), 205-218
Bryman, A., and Bell, E. (2007). Business Research Methods. New York: Oxford
University Press.
Burgess, N. and Radnor, Z. (2010). Lean Paradox: can lean influence healthcare?. In:
Proceedings of the 17th International Annual EurOMA Conference, Portugal, June,
193
2010), 1-10
Burgess, N., Radnor, Z. and Davies, R. (2009). Taxonomy of Lean in Healthcare: A
Framework for Evaluating Activity and Impact. In: Proceedings of the 16th
International Annual EurOMA Conference, Gothenburg, Sweden, June (not provided),
1-10
Bushell, S., Mobley, J. and Shelest, B. (2002). Discovering lean thinking at progressive
healthcare. Journal for Quality and Participation, 25 (2), 20-25
Cagwin, D. and Bouwman, M.J. (2002). The association between activity-based costing
and improvement in financial performance. Management Accounting Research, 13 (1),
1-39
Camacho-Miñano, M., Moyano-Fuentes, J. and Sacristán-Díaz, M. (2013). What can we
learn from the evolution of research on lean management assessment? International
Journal of Production Research, 51 (4), 1098–1116
Capon, N., Farley, J.U. and Hoenig, S. (1990). Determinants of Financial Performance: A
Meta-Analysis. Management Science, 36 (10), 1143-1159
Cappelli, P. and Neumark, D. (2001). Do "High Performance" work practices improve
establishment-level outcomes? Industrial and Labor Relations Review, 54 (4), 737-775
Carenys, J. and Sales, X. (2008). Costing the Banking Services: A Management
Accounting Approach. Journal of Money, Investment and Banking, (6), 35-53
Carnes, K. and Hedin, S. (2005). Accounting for Lean Manufacturing: Another Missed
Opportunity? Management Accounting Quarterly, 7 (1), 28-35
Carter, B., Danford, A., Howcroft, D., Richardson, H., Smith, A. and Taylor, P. (2011). All
they lack is a chain' lean and the new performance management in the British civil
service. New Technology, Work and Employment, 26 (2), 83-97
Challis, D., Samson, D. and Lawson, B. (2002). Integrated manufacturing, employee and
194
business performance: Australian and New Zealand evidence. International Journal of
Production Research, 40 (8), 1941-1964
Chaneski, WS. (2005). Stories from the Lean Office. Modern Machine Shop, 78 (7), 44-46
Chase, R. B. and Apte, U. M. (2007). A history of research in service operations: What's
the big idea? Journal of Operations Management, 25 (2), 375-386
Chenhall, R.H. (2003). Management control systems design within its organizational
context: findings from contingency-based research and directions for the future.
Accounting, Organizations and Society, 28 (2-3), 127-168
Chenhall, R.H. and Langfield-Smith, K. (1998). The relationship between strategic
priorities, management techniques and management accounting: an empirical
investigation using a systems approach. Accounting, Organizations and Society, 23 (3),
243-264
Chenhall, R.H. and Moers, F. (2007). The Issue of Endogeneity within Theory-Based,
Quantitative Management Accounting Research. European Accounting Review, 16 (1),
173-196
Chiarini, A. (2012). Lean production: mistakes and limitations of accounting systems
inside the SME sector. Journal of Manufacturing Technology Management, 23 (5),
681-700
Chin, W. W. (1998). The Partial Least Squares approach to Structural Equation Modeling.
In G. A. Marcoulides, Modern Methods for business research (pp. 295-336). New
Jersey: Lawerence Erlbaum Associates.
Chong, V.K. (1996). Management accounting systems, task uncertainty and managerial
performance: A research note. Accounting, Organizations and Society, 21 (5), 415-421
Chong, V.K. and Chong, K.M. (1997). Strategic Choices, Environmental Uncertainty and
SBU Performance: A Note on the Intervening Role of Management Accounting
195
Systems. Accounting and Business Research, 27 (4), 268-276
Christie, A.A., Joye, M.P. and Watts, R.L. (2003). Decentralization of the firm: theory and
evidence. Journal of Corporate Finance, 9 (1), 3-36
Christopher, M. (2000). The agile supply chain competing in volatile markets. Industrial
Marketing Management, 29 (1), 37-44
Clarke, P. and Mullins, T. (2001). Activity based costing in the non-manufacturing sector
in Ireland: A preliminary investigation. Irish Journal of Management, 22 (2), 1-18
Coad, A., Segarra, A. and Teruel, M. (2013). Like milk or wine: Does firm performance
improve with age? Structural Change and Economic Dynamics, 24, 173– 189
Collis, J., and Hussey, R. (2009). Business Research: A Practical Guide for Undergraduate
and Postgraduate Students. Basingstoke: Palgrave Macmillan.
Comm, C. L and Mathaisel, D. F. X. (2000). A paradigm for benchmarking lean initiatives
for quality improvement. Benchmarking: An International Journal, 7 (2), 118-127
Comm, C. L and Mathaisel, D. F. X. (2005(a)). A case study in applying lean sustainability
concepts to universities. International Journal of Sustainability in Higher Education, 6
(2), 134-146
Comm, C. L and Mathaisel, D. F. X. (2005(b)). An exploratory study of best lean
sustainability practices in higher education. Quality Assurance in Education, 13 (3),
227-240
Comm, C. L. (1999). Marketing Lean Initiatives in Service Industries. Journal of
Professional Service Marketing, 18 (2), 59-64
Cooper, R. and Kablan, R. (1992). Activity-Based Systems: Measuring the Costs of
Resource Usage. Accounting Horizons, 6 (3), 2-13
Cooper, R. and Kaplan, R.S. (1988). Measure costs right: make the right decisions.
196
Harvard Business Review, 66 (5), 96-103
Cooper, R. G. and Mohabeersingh, C. (2008(a)). Lean thinking for medical practices.
Journal of Pre-Clinical and Clinical Research, 2 (1), 88-89
Cooper, R. G. and Mohabeersingh, C. (2008(b)). Lean thinking in a healthcare system–
innovative roles. Journal of Pre-Clinical and Clinical Research, 2 (2), 110-117
Cooper,R.and Maskell, B. (2008). How to Manage Through Worse-Before-Better. MIT
Sloan management review , 49 (4), 58-65
Cortina, J. (1993). What is coefficient alpha? An examination of theory and methods.
Journal of Applied Psychology, 78 (1),98-104
Craighead, C.W., Hult, G.T.M. and Ketchen Jr, D.J. (2009). The effects of innovation–cost
strategy, knowledge, and action in the supply chain on firm performance. Journal of
Operations Management, 27 (5), 405-421
Cua, K.O., McKone, K.E. and Schroeder, R.G. (2001). Relationships between
implementation of TQM, JIT, and TPM and manufacturing performance. Journal of
Operations Management, 19 (6), 675–694
Cuatrecasas, L. (2004). A lean management implementation method in service operations.
Int. J. Services Technology and Management, 5 (5-6), 532-544
Cusumano, M. A. (1994). The Limits of "Lean". Sloan Management Review, 35 (4), 27-38
Dabhilkar, M. and Åhlström, P. (2013). Converging Production Models: The STS Versus
Lean Production Debate Revisited. International Journal of Operations and Production
Management, 33 (8), 1019-1039
Dahlgaard, J.J. and Dahlgaard-Park, S.M. (2006). Lean production, six sigma quality,
TQM and company culture. The TQM Magazine, 18 (3), 263-281
Dankbaar, B. (1997). Lean production: denial, confirmation or extension of sociotechnical
197
systems design. Human Relations, 50 (5), 567-583
Das, A. and Jayaram, J. (2007). Socio-technical perspective on manufacturing system
synergies. International Journal of Production Research, 45 (1), 169-205
Datar, S., Kekre, S., Mukhopadaya, M. and Svaan, E. (1991). Overloaded Overheads:
Activity-Based Cost Analysis of Material Handling in Cell Manufacturing. Journal of
Operations Management, 10 (1), 119-137
Davidson, A.R. and Stern, L.W. (2004). A quality self-assessment model. Managerial
Auditing Journal, 19 (7), 859-868
De Leeuw, S. and van den Berg, J.P. (2011). Improving operational performance by
influencing shopfloor behavior via performance management practices. Journal of
Operations Management, 29 (3), 224-235
De Menezes, L.M. and Wood, S. (2006). The reality of flexible work systems in Britain.
The International Journal of Human Resource Management, 17 (1), 106-138
De Menezes, L.M., Wood, S. and Gelade, G. (2010). The integration of human resource
and operation management practices and its link with performance: A longitudinal
latent class study. Journal of Operations Management, 28 (6), 455-471
Dearden, J. (1978). Cost accounting comes to service industries. Harvard Business Review,
56 (5), 132-140
Delaney, J.T. and Huselid, M.A. (1996). The impact of human resource management
practices on perceptions of organizational performance. Academy of Management
Journal, 39 (4), 949-969
Dickson, E. W., Anguelov, Z., Vetterick, D., Eller, A. and Singh. S. (2009). Use of Lean in
the Emergency Department: A Case Series of 4 Hospitals. Annals of Emergency
Medicine, 54 (4), 504-510
Dillman, D. A. (2007). Mail and Internet Surveys: The Tailored Design Method, second
198
edition. Hoboken, New Jersey: John Wiley and Sons, Inc.
Doman, M.S. (2011). A new lean paradigm in higher education: a case study. Quality
Assurance in Education, 19 (3), 248-262
Drazin, R. and Van de Ven, A.H. (1985). Alternative forms of fit in contingency theory.
Administrative Science Quarterly, 30 (4), 514-539
Drury, C. and Tayles, M. (2005). Explicating the design of overhead absorption procedures
in UK organizations. The British Accounting Review, 37 (1), 47–84
Ehrlich, B. H. (2006). Service with a smile: lean solutions extend beyond the factory floor.
Industrial Engineer, 38 (8), 40-44
Ellram, L.M., Tate, W.L. and Billington, C. (2004). Understanding and Managing the
Services Supply Chain. Journal of Supply Chain Management, 40 (4), 17-32
Emery, F.E. and Trist, E.L. (1965). The Causal Texture of Organizational Environments.
Human relations, 18, 21-32
Emiliani, M. L. (2004). Improving business school courses by applying lean principles and
practices. Quality Assurance in Education, 12 (4), 175-187
Endsley, S., Magill, M. K. and Godfrey, M. M. (2006). Creating a lean practice. Family
Practice Management, 13 (4), 34-38
Esain, A., Williams, S. and Massey, L. (2008). Combining Planned and Emergent Change
in a Healthcare Lean Transformation. Public Money and Management, 28 (1), 21-26
Fillingham, D. (2007). Can lean save lives? Leadership in Health Services, 20 (4), 231-241
Finigan, T. and Humphries, J. (2006). Maintenance gets lean. Industrial Engineer, 38 (10),
26-31
Flynn, B.B., Sakakibara, S. and Schroeder, R.G. (1995). Relationship between JIT and
199
TQM: practices and performance. The Academy of Management Journal, 38 (5), 1325-
1360
Fornell, C. G., and Larcker, D. F. (1981). Evaluating structural equation models with
unobservable variables and measurement error. Journal of Marketing Research, 18(1),
39-50
Fox, W.M. (1995). Sociotechnical System Principles and Guidelines: Past and Present.
Journal of Applied Behavioral Science, 31 (1), 91-105
Frazier, P.A., Tix, A.P. and Barron, K.E. (2004). Testing Moderator and Mediator Effects
in Counseling Psychology Research. Journal of Counseling Psychology, 51 (1), 115-
134
Frey, K. and Gordon, L.A. (1999). ABC, strategy and business unit performance.
International Journal of Applied Quality Management, 2 (1), 1-23
Fullerton, R. R. and McWatters, C. S. (2002). The role of performance measures and
incentive systems in relation to the degree of JIT implementation. Accounting,
Organizations and Society, 27 (8), 711–735
Fullerton, R. R. and Wempe, W. F. (2009). Lean manufacturing, non-financial
performance measures, and financial performance. International Journal of Operations
and Production Management, 29 (3), 214-240
Fullerton, R. R., McWatters, C. S. and Fawson, C. (2003). An examination of the
relationships between JIT and financial performance. Journal of Operations
Management, 21 (4), 383-404
Fullerton, R.R. and McWatters, C.S. (2004). An empirical examination of cost accounting
practices used in advanced manufacturing environments. Advances in Management
Accounting, 12 (-), 85–113
Fullerton, R.R., Kennedy, F.A. and Widener, S.A. (2013). Management accounting and
200
control practices in a lean manufacturing environment. Accounting, Organizations and
Society, 38 (1), 50–71
Furlan, A., Vinelli, A. and Dal Pont, G. (2011). Complementarity and lean manufacturing
bundles: an empirical analysis. International Journal of Operations and Production
Management, 31 (8), 835-850
Galende, J. and de la Fuente, J.M. (2003). Internal factors determining a firm’s innovative
behaviour. Research Policy, 32 (5), 715 - 736
Garrett, D.F. and Lee, J. (2010). Lean Construction Submittal Process-A Case Study.
Quality Engineering, 23 (1), 84-93
Geisser, S. (1974). A Predictive Approach to the Random Effects Model. Biometrika, 61
(1), 101–107.
Gerdin, J. (2005). The Impact of departmental interdependencies and management
accounting system use on subunit performance. European Accounting Review, 14 (2),
297-327
Glancey, K. (1998). Determinants of growth and profitability in small entrepreneurial
firms. International Journal of Entrepreneurial Behaviour & Research, 4 (1), 18-27
González-Benito, J. (2005). A study of the effect of manufacturing proactivity on business
performance. International Journal of Operations and Production Management, 25 (3),
222-241
Gorman, M.F., Hoff, J. and Kinion, R. (2009). ASP, The Art and Science of Practice: Tales
from the Front: Case Studies Indicate the Potential Pitfalls of Misapplication of Lean
Improvement Programs. Interfaces, 39 (6), 540-548
Gosselin, M. (1997). The effect of strategy and organizational structure on the adoption
and implementation of activity-based costing. Accounting, Organizations and Society,
22 (2), 105–122
201
Grafton, J., Lillis, A.M. and Widener, S.K. (2010). The role of performance measurement
and evaluation in building organizational capabilities and performance. Accounting,
Organizations and Society, 35 (7), 689–706
Grasso, L. (2005). Are ABC and RCA Accounting Systems Compatible with Lean
Management? Management Accounting Quarterly, 7 (1), 12-27
Guest, D.E., Michie, J., Conway, N. and Sheehan, M. (2003). Human Resource
Management and Corporate Performance in the UK. British Journal of Industrial
Relations, 41 (2), 291–314
Guilding, C., Drury, C. and Tayles, M. (2005). An empirical investigation of the
importance of cost-plus pricing. Managerial Auditing Journal, 20 (2), 125-137
Gunasekaran, A. and Sarhadi, M. (1998). Implementation of activity-based costing in
manufacturing. International Journal of Production Economics, 56-57 (20), 231-242
Hagan, P. (2011). Waste not, want not Leading the lean health-care journey at Seattle
Children's Hospital. Global Business and Organizational Excellence, 30 (3), 25-31
Hair, J.F., Ringle, C.M. and Sarstedt, M . (2011). PLS-SEM: Indeed a silver bullet. Journal
of Marketing Theory and Practice, 19 (2), 139–151
Hair, J.F., Ringle, C.M. and Sarstedt, M . (2013). Partial least squares structural equation
modeling: rigorous applications, better results and higher acceptance. Long Range
Planning, 46 (1-2), 1–12
Hair, J.F., Sarstedt, M. and Ringle, C.M. (2012). An assessment of the use of partial least
squares structural equation modeling in marketing research. Journal of the Academy of
Marketing Science, 40 (3), 414-433
Hair, Joseph F, William C. Black, Barry J.Babin, and Rolph E. Anderson, (2010).
Multivariate Data Analysis. A Global Perspective, 7th edition. New Jersey: Pearson
Education,Inc
202
Hameri, AP. (2010). Production flow analysis-Cases from manufacturing and service
industry. International Journal of Production Economics, 129 (2), 233-241
Hammad, S.A., Jusoh, R. and Oon, E.Y.N. (2010). Management accounting system for
hospitals: a research framework. Industrial Management and Data Systems, 110 (5),
762-784
Hannan, M.T. and Freeman, J. (1984). Structural Inertia and Organizational Change.
American Sociological Review, 49 (2), 149-164
Haque, B. and James-Moore, M. (2004). Applying lean thinking to new product
introduction. Journal of Engineering Design, 15 (1), 1-31
Harris, L.G. (2001). Market Orientation and Performance: Objective and Subjective
Empirical Evidence from UK Companies. Journal of Management Studies, 38 (1), 17-
43
Hartwell, J. and Roth, G. (2010). Doing more with less at Ariens: a leadership and
transformation case study. Organization Management Journal, 7 (2), 89-109
Hegde, G.G. and Nagarajan, N.J. (1992). Incentives for overhead cost reduction: Setup
time and lot size considerations. International Journal of Production Economics, 28 (3),
255-263
Henri, J. -F. (2007). A quantitative assessment of the reporting of structural equation
modeling information. Journal of Accounting Literature, 76-115.
Henseler, J., and Chin, W. W. (2010). A comparison of approaches for the analysis of
interaction effects between latent variables using Partial Least Squares path modeling.
Structural Equation Modeling: A miltidiscipilnary journal, 17(1), 82-109.
Hillman, A.J. (2005). Politicians on the Board of Directors: Do Connections Affect the
Bottom Line? Journal of Management, 31 (3), 464-481
Hines, P. and Lethbridge, S. (2008). New Development: Creating a Lean University.
203
Public Money and Management, 28 (1), 53-56
Hines, P., Holweg, M. and Rich, N. (2004). Learning to evolve: A review of contemporary
lean thinking. International Journal of Operations and Production Management, 24
(10), 994-1011
Hines, P., Martins, A. L. and Beale, J. (2008). Testing the Boundaries of Lean Thinking:
Observations from the Legal Public Sector. Public Money and Management, 28 (1), 35-
40
Hinterhuber, H. H. (1994). The European Way to Lean Management. The International
Executive, 36 (3), 275-290
Holden, R. J. (2010). Lean Thinking in Emergency Departments: A Critical Review.
Annals of Emergency Medicine, 57 (3), 265-278
Holm, M. and Ahlstrom, P. (2010a). Assessing lean service integration. In: Proceedings of
the 17th International Annual EurOMA Conference, Portugal, June, 1-10
Holm, M. and Ahlstrom, P. (2010b). Lean Service - a literature review. In: Proceedings of
the 17th International Annual EurOMA Conference, Portugal, June, 1-10
Holweg, M. (2007). The genealogy of lean production. Journal of Operations
Management, 25 (2), 420-437
Höök, M. and Stehn, L. (2008). Lean principles in industrialized housing production: the
need for a cultural change. Lean Construction Journal, 20-33
Huber, V.L. and Brown, K.A. (1991). Human resource issues in cellular manufacturing: A
sociotechnical analysis. Journal of Operations Management, 10 (1), 138-159
Huselid, M.A. (1995). The impact of human resource management practices on turnover,
productivity, and corporate financial performance. Academy of Management Journal,
38 (3), 635-672
204
Hyer, N.L. and Wemmerlöv, U. (2002). The office that lean built. Industrial Management,
44 (5), 36-43
Inman, R.A., Sale, R.S., Green Jr, K.W. and Whitten, D. (2011). Agile manufacturing:
Relation to JIT, operational performance and firm performance. Journal of Operations
Management, 29 (4), 343–355
Innes, J. and Mitchell, F. (1995). A survey of activity-based costing in the U.K.'s largest
companies. Management Accounting Research, 6 (2), 137-153
Ittner, C. D. (1999). Activity-based costing concepts for quality improvement. European
Management Journal, 17 (5), 492-500
Ittner, C. D., and Larcker, D. F. (2001). Assessing empirical research in managerial
accounting: A value-based management perspective. Journal of Accounting and
Economics, 32, 349-410.
Ittner, C.D., Lanen, W.N. and Larcker, D.F. (2002). The Association Between Activity-
Based Costing and Manufacturing Performance. Journal of Accounting Research, 40
(3), 711-726
Jaaron, A. and Backhouse, C.J. (2011). A methodology for the implementation of lean
thinking in manufacturing support services. Int. J. Services and Operations
Management, 9 (4), 389-410
Jain, R. and Lyons, A.C. (2009). The implementation of lean manufacturing in the UK
food and drink industry. Int. J. Services and Operations Management, 5 (4), 548-573
Jayaram, J., Ahire, S.L. and Dreyfus, P. (2010). Contingency relationships of firm size,
TQM duration, unionization, and industry context on TQM implementation—A focus
on total effects. Journal of Operations Management, 28 (4), 345–356
Jimmerson, C., Weber, D. and Sobek, D. K. (2005). Reducing Waste and Errors: Piloting
Lean Principles at IHC. Joint Commission Journal on Quality and Patient Safety, 31
205
(5), 249-257
Jones, C., Medlen, N., Merlo, C., Robertson, M. and Shepherdson, J. (1999). The lean
enterprise. BT Technology Journal, 17 (4), 15-22
Joosten, T., Bongers, I. and Jansses, R. (2009). Application of lean thinking to health care:
Issues and observations. International Journal for Quality in Health Care, 21 (5), 341-
347
Joshi, M.P., Kathuria, R. and Porth, S.J. (2003). Alignment of strategic priorities and
performance: an integration of operations and strategic management perspectives.
Journal of Operations Management, 21 (3), 353–369
Julien, D.M. and Tjahjono, B. (2009). Lean thinking implementation at a safari park.
Business Process Management Journal, 15 (3), 321-335
Kallunki, JP. and Silvola, H. (2008). The effect of organizational life cycle stage on the use
of activity-based costing. Management Accounting Research, 19 (1), 62–79
Kaplan, G.S. and Patterson, S.H. (2008). Seeking perfection in Healthcare: A Case Study
in Adopting Toyota Production System Methods. Healthcare Executive, 23 (3), 17-20
Karlsson, C. and Ahlstrom, P. (1996). Assessing changes towards lean production.
International Journal of Operations and Production Management, 16 (2), 24-41
Karmarkar, U. (2004). Will you survive the services revolution? Harvard Business Review,
82 (6), 100-107
Kaynak, H. (2003). The relationship between total quality management practices and their
effects on firm performance. Journal of Operations Management, 21 (4), 405-435
Keen, M. (2011). Office skills. Management Services, 55 (1), 20-22
Kennedy, F., Owens-Jackson, L., Burney, L. and Schoon, M. (2007). How do your
206
measurements stack up to lean? Strategic Finance, 88 (11), 33-41
Kennedy, F.A. and Widener, S.K. (2008). A control framework: insights from evidence on
lean accounting. Management Accounting Research, 19 (4), 301-323
Kennedy, T. and Affleck-Graves, J. (2001). The impact of activity-based costing
techniques on firm performance. Journal of Management Accounting Research, 13, 19-
45
Kepes, S., Banks, G.C., McDaniel, M. and Whetzel, D.L. (2012). Publication Bias in the
Organizational Sciences. Organizational Research Methods, 15 (4), 624-662
Khataie, A.H. and Bulgak, A.A. (2013). A cost of quality decision support model for lean
manufacturing: activity-based costing application. International Journal of Quality &
Reliability Management, 30 (7), 751-764
Khataie, A.H., Bulgak, A.A and Segovia, J.J. (2011). Activity-Based Costing and
Management applied in a hybrid Decision Support System for order management.
Decision Support Systems, 52 (1), 142–156
Kim, C. S., Spahlinger, D. A., Kin, J. M. and Billi, J. A. (2006). Lean Health Care: What
Can Hospitals Learn from a World-Class Automaker? Journal of Hospital Medicine, 1
(3), 191-199
Kim, D.U., Kumar, V. and Kumar, U. (2012). Relationship between quality management
practices and innovation. Journal of Operations Management, 30 (4), 295-315
King, D. L., Ben-Tovim, D. I. and Bassham, J. (2006). Redesigning emergency department
patient flows: Application of Lean Thinking to health care. Emergency Medicine
Australasia, 18 (4), 391-397
Kock, S. (1995). Implementation considerations for activity-based cost systems in service
firms: The unavoidable challenge. Management Decision, 33 (6), 57-63
Kosuge, R., Modig, N. and Ahlstrom, P. (2010). Standardization in lean service: Exploring
207
the contradiction. In: Proceedings of the 17th International Annual EurOMA
Conference, Portugal, June, 2010), 1-10
Krafcik, J. F. (1988). Triumph of the lean production system. Sloan Management Review,
30 (1), 41-52
Kress, N.J. (2008). Lean Thinking in Libraries: A Case Study on Improving Shelving
Turnaround. Journal of Access Services, 5 (1-2), 159-172
Kroes, J.R. and Ghosh, S. (2010). Outsourcing congruence with competitive priorities:
Impact on supply chain and firm performance. Journal of Operations Management, 28
(2), 124–143
Kull, T.J., Ellis, S.C. and Narasimhan, R. (2013). Reducing Behavioral Constraints to
Supplier Integration: A Socio-Technical Systems Perspective. Journal of Supply Chain
Management, 49 (1),64-86
Kumar, A. and Telang, R. (2011). Product Customization and Customer Service Costs: An
Empirical Analysis. Manufacturing and Service Operations Management, 13 (3), 347-
360
Kuriger, G.W., Wan, H.D., Mirehei, M., Tamma, S. and Chen, F.F. (2010). A Web-Based
Lean Simulation Game for Office Operations: Training the Other Side of a Lean
Enterprise. Simulation and Gaming, 41 (4), 487-510
LaGanga, L.R. (2011). Lean service operations: Reflections and new directions for
capacity expansion in outpatient clinics. Journal of Operations Management, 29 (5),
422-433
Lamminmaki, D. and Drury, C. (2001). A comparison of New Zealand and British
product-costing practices. The International Journal of Accounting, 36 (3), 329–347
Langfield-Smith, K. (1997). Management control systems and strategy: A critical review.
Accounting, Organizations and Society, 22 (2), 207-232
208
Larcker, D.F. and Rusticus, T.O. (2007). Endogeneity and Empirical Accounting Research.
European Accounting Review, 16 (1), 207-215
Larson, P.D. and Kerr, S.G. (2007). Integration of Process Management Tools to Support
TQM Implementation: ISO 9000 and Activity-based Costing. Total
Quality Management and Business Excellence, 18 (1), 201-207
Lasa, I.S., Castro, R. and Laburu, C.O. (2009). Extent of the use of Lean concepts
proposed for a value stream mapping application. Production Planning and Control, 20
(1), 82-98
Lazarus, I. R. and Andell, J. (2006). Providers, payers and it suppliers learn it pays to get
"Lean". Managed Healthcare Executive, 16 (2), 34-36
Lee, L., Petter, S., Fayard, D. and Robinson, S. (2011). On the use of partial least squares
path modeling in accounting research. International Journal of Accounting Information
Systems, 12 (4), 305-328
Lee, S. M., Olson, D. L., Lee, S. H., Hwang, T. and Shin, M. S. (2008). Entrepreneurial
applications of the lean approach to service industries. The Service Industries Journal,
28 (7), 973-987
Lei, D., Hitt, M.M. and Goldhar, J.D. (1996). Advanced Manufacturing Technology:
Organizational Design and Strategic Flexibility. Organization Studies, 17 (3), 501-523
Lewis, M.A. (2000). Lean production and sustainable competitive advantage. International
Journal of Operations and Production Management, 20 (8), 959-978
Li, X., Sawhney, R., Arendt, E.J. and Ramasamy, K. (2012). A comparative analysis of
management accounting systems' impact on lean implementation. International Journal
of Technology Management, 57 (1-3), 33-48
Liker, J. K. and Morgan, J. M. (2006). The Toyota Way in Services: The Case of Lean
Product Development. Academy of Management Perspectives, 20 (2), 5-20
209
Lodge, A. and Bamford, D. (2008). New Development: Using Lean Techniques to Reduce
Radiology Waiting Times. Public Money and Management, 28 (1), 49-52
Lowry, J.F. (1990). Management Accounting and Service Industries: An Exploratory
Account of Historical and Current Economic Contexts. ABACUS, 26 (2), 159-184
Luft, J. and Shields, M. (2003). Mapping management accounting graphics and guidelines
for theory-consistent empirical research. Accounting, Organizations and Society, 28 (2-
3), 169–249
Lundvall, K. and Battese, G.E. (2000). Firm size, age and efficiency: Evidence from
Kenyan manufacturing firms. The Journal of Development Studies, 36 (3), 146-163
Machuca, J. A. D., Gonza´lez-Zamora, M. d. M. and Aguilar-Escobar, V. G. (2007).
Service operations management research. Journal of Operations Management, 25 (3),
585-603
MacKinnon, D.P., Krull, J.L. and Lockwood, C.M. (2000). Equivalence of the mediation,
confounding and suppression effect. Prevention Science, 1 (4), 173-181
Maguad, BA. (2007). Lean Strategies for Education: Overcoming the Waste Factor.
Project Innovation, 128 (2), 248-255
Maiga, A.S. and Jacobs, F.A. (2008). Extent of ABC Use and Its Consequences.
Contemporary Accounting Research, 25 (2), 533-566
Malladi, S., Dominic, P.D.D. and Kamil, A. (2010). Lean principles in IT services: a case
study on implementation and best practices. Int. J. Business Information Systems, 8 (3),
1-22
Malmbrandt, M., & Åhlström, P. (2013). An Instrument for Assessing Lean Service
Adoption. International Journal of Operations and Production Management, 33 (9),
1131-1165
Manos, A., Sattler, M. and Alukal, G. (2006). Make Healthcare Lean. Quality Progress, 39
210
(7), 24-30
Manz, C.C. and Stewart, G.L. (1997). Attaining Flexible Stability by Integrating Total
Quality Management and Socio-Technical Systems Theory. Organization Science, 8
(1), 59-70
Martinson, O.B. (2002). A look at cost accounting in the service industry and the federal
government. The Journal of Government Financial Management, 51 (1), 18-25
Maskell, B.H. (2006). Solving the standard cost problem. Cost Management, 20 (1), 27-35
Maskell, B.H. and Kennedy, F.A. (2007). Why do we need lean accounting and how does
it work? Journal of Corporate Accounting and Finance, 18 (3), 59-73
McQuade, D. (2008). New Development: Leading Lean Action to Transform Housing
Services. Public Money and Management, 28 (1), 57-60
Meade, D. J., Kumar, S. and White, B. (2010). Analysing the impact of the implementation
of lean manufacturing strategies on profitability. Journal of the Operational Research
Society, 61 (5), 858-871
Mefford, R. N. (1993). Improving service quality: Learning from manufacturing.
International Journal of Production Economics, 30-31, July, 399-413
Meier, H. S. and Forrester, P. L. (2002). A model for evaluating the degree of leanness of
manufacturing firms. Integrated Manufacturing Systems, 13 (2), 104-109
Mia, L. and Clarke, B. (1999). Market competition, management accounting systems and
business unit performance. Management Accounting Research, 10 (2), 137-1 58
Middleton, P., Flaxel, A. and Cookson, A. (2005). Lean Software Management Case
Study: Timberline Inc.. Lecture Notes in Computer Science, 3556, 1-9
Miles, R. E., Snow, C.C., Meyer, A.D. and Coleman, H.J. (1978). Organizational strategy,
211
structure, and process. Academy of Management Review, 3 (3), 546-562
Miller, D. (1981). Toward a new contingency approach: The search for organisational
gestals. Journal of Management Studies, 18(1), 1–26
Mirehei, S.M., Kuriger, G., Wan. H.D. And Chen, F.F. (2011). Enhancing lean training for
the office environment through simulation and gaming. Int. J. Learning and Intellectual
Capital, 8 (2), 206-221
Mishra, B. and Vaysman, I. (2001). Cost-System Choice and Incentives-Traditional vs.
Activity-Based Costing. Journal of Accounting Research, 39 (3), 619-641
Moayed, F.A. and Shell, R.L. (2009). Comparison and evaluation of maintenance
operations in lean versus non-lean production systems. Journal of Quality in
Maintenance Engineering, 15 (3), 285-296
Mohrman, S.A., Tenkasi, R.V., Lawler, E.E. and Ledford, G.E. (1995). Total quality
management: practice and outcomes in the largest US firms. Employee Relations, 17
(3), 26-41
Montabon, F. (2005). Using kaizen events for back office processes: the recruitment of
frontline supervisor co-ops. Total Quality Management and Business Excellence, 16
(10), 1139–1147
Nelson-Peterson, D.L. and Leppa, C.J. (2007). Creating an environment for caring using
lean principles of the Virginia mason production system. The journal of nursing
administration, 37 (6), 287-294
Nie, W. and Kellogg, D. L. (1999). How professors of operations management view
service operations? Production and Operations Management, 8 (3), 339-355
Nielsen, A. P. and Edwards, k. (2010). Implementing lean in healthcare: Barriers and
opportunities. In: Proceedings of the 17th International Annual EurOMA Conference,
Portugal, June, 2010), 1-10
212
Nunnally, J. (1978). Psychometric Theory, second ed. McGraw-Hill, New York.
Ohno, T. (1988). Toyota Production System: Beyond Large-Scale Production. Portland,
OR: Productivity Press.
Oliver, N., Delbridge, R., Jones, D. and Lowe, J. (1994). World Class Manufacturing:
Further Evidence in the Lean Production Debate. British Journal of Management, 5 (1),
53-63
Papadopoulos, T. and Merali, Y. (2008). Stakeholder Network Dynamics and Emergent
Trajectories of Lean Implementation Projects: A Study in the UK National Health
Service. Public Money and Management, 28 (1), 41-48
Patterson, M.G., West, M.A. and Wall, T.D. (2004). Integrated manufacturing,
empowerment and company performance. Journal of Organizational Behavior, 25 (5),
641-665
Pavlatos, O. (2010). The impact of firm characteristics on ABC systems: a Greek-based
empirical analysis. Studies in Managerial and Financial Accounting, 20, 501–527
Pavnaskar, S.J., Gershenson, J.K. and Jambekar, A.B. (2003). Classification scheme for
lean manufacturing tools. International Journal of Production Research, 41 (13), 3075-
3090
Pedersen, ERG. and Huniche, M. (2011). Negotiating lean: The fluidity and solidity of
new management technologies in the Danish public sector. International Journal of
Productivity and performance management, 60 (6), 550-566
Pejsa, P. and Eng, R. (2011). Lean strategy deployment delivers customer satisfaction at
GE Healthcare. Global Business and Organizational Excellence, 30 (5), 45-55
Peng, D.X. and Lai, F. (2012). Using partial least squares in operations management
research: A practical guideline and summary of past research. Journal of Operations
Management, 30 (6), 467-480
213
Petersen, K. and Wohlin, C. (2010). Software process improvement through the Lean
Measurement (SPI-LEAM) method. The Journal of Systems and Software, 83 (7),
1275–1287
Pettersen, J. (2009). Defining lean production: some conceptual and practical issues. The
TQM Journal, 21 (2), 127-142
Piercy, N. and Rich, N. (2009(a)). High quality and low cost: the lean service center.
European Journal of Marketing, 43 (11), 1477-1497
Piercy, N. and Rich, N. (2009(b)). Lean transformation in the pure service environment:
the case of the call service centre. International Journal of Operations and Production
Management, 29 (1), 54-76
Pizzini, M.J. (2006). The relation between cost-system design, managers’ evaluations of
the relevance and usefulness of cost data, and financial performance: an empirical study
of US hospitals. Accounting, Organizations and Society, 31 (2), 179–210
Podsakoff, P.M., Mackenzie, S.B., Lee, J.Y., & Podsakoff, N.P. (2003). Common method
biases in behavioral research: a critical review of the literature and recommended
remedies. Journal of Applied Psychology, 88 (5), 879–903
Poksinska, B. (2010). The Current State of Lean Implementation in Health Care: Literature
Review. Quality Management in Health Care, 19 (4), 319-329
Pont, G.D., Furlan, A. and Vinelli, A. (2008). Interrelationships among lean bundles and
their effects on operational performance. Operations Management Research, 1 (2), 150-
158
Porter, M. (1980). Competitive Strategy, Free Press. New York. NY.
Powell, T.C. (1995). Total Quality Management as Competitive Advantage: A Review and
Empirical Study. Strategic Management Journal, 16 (1), 15-37
Qi, Y., Zhao, X. and Sheu, C. (2011). The Impact of Competitive Strategy and Supply
214
Chain Strategy on Business Performance: The Role of Environmental Uncertainty.
Decision Sciences, 42 (2), 371-389
Qudrat-Ullah, H., Seong, B.S. and Mills, B.L. (2012). Improving high variable-low volume
operations: an exploration into the lean product development. International Journal of
Technology Management, 57 (1-23), 49-70
Radnor, Z. (2010). Transferring lean into government. Journal of Manufacturing
Technology Management, 21 (3), 411-428
Radnor, Z. and Boaden, R. (2008). Lean in Public Services – Panacea or Paradox? Public
Money and Management, 28 (1), 3-7
Radnor, Z. and Walley, P. (2008). Learning to Walk Before We Try to Run: Adapting
Lean for the Public Sector. Public Money and Management, 28 (1), 13-20
Rahman, S., Laosirihongthong, T. and Sohal, A.S. (2010). Impact of lean strategy on
operational performance: A study of Thai manufacturing companies. Journal of
Manufacturing Technology Management, 21 (7), 839-852
Rajiv, S. (2009). The skinny on lean logistics. Industrial Engineer, 41 (10), 20-20
Rashidirad, M., Soltani, E. and Syed, J. (2013). Strategic Alignment Between Competitive
Strategy and Dynamic Capability: Conceptual Framework and Hypothesis
Development. Strategic Change, 22 (3-4), 213-224
Reinertsen, D. (2005). Let it flow. Industrial Engineer, 37 (6), 40-45
Reinertsen, D. and Shaeffer, L. (2005). Making R&D lean. Research Technology
Management, 48 (4), 51-57
Rexhausen, D., Pibernik, R. and Kaiser, G. (2012). Customer-facing supply chain
practices—The impact of demand and distribution management on supply chain
success. Journal of Operations Management, 30 (4), 269–281
215
Robinson, A. G. and Schroeder, D. M. (2009). The Role of Front-Line Ideas in Lean
Performance Improvement. The Quality Management Journal, 16 (4), 27-40
Ruiz-de-Arbulo-Lopez, P., Fortuny-Santos, J. and Cuatrecasas-Arbo´s, L. (2013). Lean
manufacturing: costing the value stream. Industrial Management and Data Systems,
113 (5), 647-668
Saidur. R., Atabani, A.E. and Mekhilef, S. (2011). A review on electrical and thermal
energy for industries. Renewable and Sustainable Energy Reviews, 15 (4), 2073–2086
Sakakibara, S., Flynn, B.B., Schroeder, R.G. and Morris, W.T. (1997). The Impact of Just-
In-Time Manufacturing and Its Infrastructure on Manufacturing Performance.
Management Science, 43 (9), 1246-1257
Sampson, S.E. and Froehle, C. M. (2006). Foundations and Implications of a Proposed
Unified Services Theory. Production and Operations Management, 15 (2),329–343
Samson, D. and Terziovski, M. (1999). The relationship between total quality management
practices and operational performance. Journal of Operations Management, 17 (4),
393-409
Sasser, W.E. (1976). Match supply and demand in service industries. Harvard Business
Review, 54 (6), 133-140
Saunders, M., Lewis, P., and Thornhill, A. (2009). Research Methods for Business
Students. London: FT Prentice Hall.
Scherrer-Rathje, M., Boyle, T.A. and Deflorin, P. (2009). Lean, take two! Reflections from
the second attempt at lean implementation. Business Horizons, 52 (1), 79-88
Schlissel, M.R. and Chasin, J. (1991). Pricing of services: an interdisciplinary review. The
Service Industries Journal, 11 (3), 271-286
Schmenner, R.W. (1986). How Can Service Businesses Survive and Prosper? Sloan
216
Management Review, 27 (3), 32
Schonberger, R. J. (2008). Lean performance management (metrics do not add up). Cost
Management, 22 (1), 5-10
Schoute, M. (2011). The relationship between product diversity, usage of advanced
manufacturing technologies and activity-based costing adoption. The British Accounting
Review, 43 (2), 120–134
Scott, W.M. and Walton, D.A. (2010). Maximizing Case Efficiency: Lessons Learned
From Lean-A Process Management Philosophy Utilized In Automotive Manufacturing.
FDCC Quarterly, Fall, 1-16
Searcy, D.L. (2009(b)). Using cost management and lean tools to improve AMG's rental
operations. Cost Management, 23 (6), 24-33
Shafer, S.M. and Moeller, S.B. (2012). The effects of Six Sigma on corporate
performance: An empirical investigation. Journal of Operations Management, 30 (7-8),
521-532
Shah, R. and Goldstein, S.M. (2006). Use of structural equation modeling in operations
management research: Looking back and forward. Journal of Operations Management,
24(2), 148-169
Shah, R. and Ward, P. T. (2003). Lean manufacturing: context, practice bundles, and
performance. Journal of Operations Management, 21 (2), 129-149
Shah, R. and Ward, P. T. (2007). Defining and developing measures of lean production.
Journal of Operations Management, 25 (4), 785-805
Shahrukh, I. (2011). Choosing what works: high-mix, low-volume small manufacturers
must supplement the right tools with other strategies. Industrial Engineer, 43 (8), 42-47
Sharma, S., Durand, R.M. and Gur-Arie, O. (1981). Identification and Analysis of
217
Moderator Variables. Journal of Marketing Research, 18 (3), 291-300
Sheu, H.J. and Pan, CY. (2009). Cost-system choice in a multidimensional knowledge
space: traditional versus activity-based costing. International Journal of Technology
Management, 48 (3), 358-388
Shields, J., and Shields, M. (1998). Antecedents of participative budgeting. Accounting,
Organizations and Society, 23, 49-76.
Shields, M. D. (1995). An empirical analysis of firms' implementation experiences with
activity-based costing. Journal of Management Accounting Research, 7 (Fall), 1-28
Shields, M.D; Deng, F.G., Kato, Y. (2000). The design and effects of control systems: tests
of direct- and indirect effects models. Accounting, Organizations and Society, 25( 2),
185-202
Smith, D., and Langfield-Smith, K. (2004). Structural equation modeling in management
accounting research: Critical analysis and opportunities. Journal of Accounting
Literature, 23, 49-86.
Soltani, E., Barnes, B., Syed, J. and Liao, Y. (2012). Does management's approach impede
service quality? Production Planning & Control: The Management of Operations, 23
(7), 523-540
Soltani, E., Liao, Y., Singh, A. and Wang, W. (2010). Managing service quality: The
managers' orientations and their consequences – case study evidence of current practice.
Total Quality Management & Business Excellence, 21(6), 673-685
South, J.B. (1993). A modified standard cost-accounting system can generate valid product
costs. Production and Inventory Management Journal, 34 (2), 28-31
Souza, L. B. (2009). Trends and approaches in lean healthcare. Leadership in Health
Services, 22 (2), 121-139
Spedding, T.A. and Sun, G.Q. (1999). Application of discrete event simulation to the
218
activity based costing of manufacturing systems. International Journal of Production
Economics, 58 (3), 289–301
Sprigg, C. A. and Jackson, P. R. (2006). Call Centers as Lean Service Environments: Job-
Related Strain and the Mediating Role of Work Design. Journal of Occupational Health
Psychology, 11 (2), 197-212
Staats, B.R. and Upton, D.M. (2011). Lean knowledge work. Harvard Business Review, 89
(10), 100-110
Staats, B.R., Brunner, D.J. and Upton, D.M. (2011). Lean principles, learning, and
knowledge work: Evidence from software services provider. Journal of Operations
Management, 29 (5), 376-390
Stone, M. (1974). Cross-Validatory Choice and Assessment of Statistical Predictions.
Journal of the Royal Statistical Society, 36 (2), 111–147.
Stuebs, M. and Sun, L. (2010). Business Reputation and Labor Efficiency, Productivity,
and Cost. Journal of Business Ethics, 96 (2), 265-283
Suarez-Barraza, M. F., Smith, T. and Dahlgaard-Park, S. M. (2009). Lean-kaizen public
service: an empirical approach in Spanish local governments. The TQM Journal, 21 (2),
143-167
Suarez-Barraza, M.F. and Ramis-Pujol, J. (2010). Implementation of Lean-Kaizen in the
human resource service process: A case study in a Mexican public service organisation.
Journal of Manufacturing Technology Management, 21 (3), 388-410
Suárez-Barraza, M.F., Smith, T. and Dahlgaard-Park, S.M. (2012). Lean Service: A
literature analysis and classification. Total Quality Management and Business
Excellence, 23 (3-4), 359-380
Swank, C. K. (2003). The Lean Service Machine. Harvard Business Review, 81 (10), 123-
129
219
Swink, M. and Jacobs, B.W. (2012). Six Sigma adoption: Operating performance impacts
and contextual drivers of success. Journal of Operations Management, 30 (6), 437-453
Tabachnick, B.G., & Fidell, L.S. (2007), Using Multivariate Statistics (5th ed.). New York:
Allyn and Bacon.
Talib, F., Rahman, Z. and Qureshi, M.N. (2013). An empirical investigation of relationship
between total quality management practices and quality performance in Indian service
companies. International Journal of Quality and Reliability Management, 30 (3), 280-
318
Tatikonda, L. (2007). Applying Lean Principles to Design, Teach, and Assess Courses.
Management Accounting Quarterly, 8 (3), 27-38
Taylor, A. and Taylor, M. (2009). Operations management research: contemporary themes,
trends and potential future directions. International Journal of Operations and
Production Management, 29 (12), 1316-1340
Taylor, A.B., MacKinnon, D.P. and Tein, J. (2008). Tests of the three-path mediated
effect. Organizational Research Methods, 11 (2), 241-269
Temme, D., Henning, K., and Hildebrandt, L. (2010). A comparison of current PLS path
modeling software: Features, ease-of-Use, and performance. In V. E. Vinzi, W.
Tiplady, B. (2010). Running a lean office. British Journal of Administrative Management,
fall (72), 22-23
Tischler, L. (2006). Bringing Lean to the Office. Quality Progress, 39 (7), 32-38
Tonya, V. (2004). Spreading The Good Word. Industry Week, 253 (2), 59-60
Toomey, J.W. (1994). Adjusting cost management systems to lean manufacturing
environments. Production and Inventory Management Journal, 35 (3), 82-85
Towne, J. (2006). Going ‘lean’ streamlines processes, empowers staff and enhances care.
220
Hospitals and Health Networks, 80 (10), 34-35
Trist, E.L. (1981), The evolution of socio-technical systems: a conceptual framework and
action research program. (Occasional paper No. 2), Ontario Quality of Working Life
Centre, Ontario.
Trist, E.L. and Bamforth, K.W. (1951). Some Social and Psychological Consequences of
the Longwall Method. Human relations, 4, 3-38
Tsai, W. and Lai, C. (2007). Outsourcing or capacity expansions: Application of activity-
based costing model on joint products decisions. Computers and Operations Research,
34 (12), 3666–3681
Turesky, E.F. and Connell, P. (2010). Off the rails: understanding the derailment of a lean
manufacturing initiative. Organization Management Journal, 7 (2), 110-132
Turney, P.B.B. and Anderson, B. (1989). Accounting For Continuous Improvement. MIT
Sloan Management Review, 30 (2), 37-47
Van Biema, M. and Greenwald, B. (1997). Managing our way to higher service-sector
productivity. Harvard Business Review, 75 (4), 87-95
Venkatraman, N. (1989). The concept of fit in strategy research: Toward verbal and
statistical correspondence. Academy of Management Review, 14 (3), 423-444
Villarreal, B., Garcia, D. and Rosas, I. (2009). Eliminating Transportation Waste in Food
Distribution: A Case Study. Transportation Journal, 48 (4), 72-77
Vlachos, I. and Bogdanovic, A. (2013). Lean thinking in the European hotel industry.
Tourism Management, 36 (June), 354-363
Voss, C., Tsikriktsis, N. and Frohlich, M. (2002). Case research in operations management.
International Journal of Operations & Production Management, 22 (2), 195-219
Voss, C.A. (2005). Paradigms of manufacturing strategy re-visited. International Journal
221
of Operations and Production Management, 25 (12), 1223-1227
Wagner, S.M., Grosse-Ruyken, P.T. and Erhun, F. (2012). The link between supply chain
fit and financial performance of the firm. Journal of Operations Management, 30 (4),
340-353
Ward, P.T. and Duray, R. (2000). Manufacturing strategy in context: environment,
competitive strategy and manufacturing strategy. Journal of Operations Management,
18 (2), 123–138
Ward, P.T., Bickford, D.J. and Leong, G.K. (1996). Configurations of Manufacturing
Strategy, Business Strategy, Environment and Structure. Journal of Management, 22
(4), 597-626
Ward, P.T., McCreery, J.K. and Anand, G. (2007). Business strategies and manufacturing
decisions: An empirical examination of linkages. International Journal of Operations
and Production Management, 27 (9), 951-973
Wayne, C. (2005). Company Applies Lean Techniques in The Office. Modern Machine
Shop, 78 (7), 44-46
Wenchao Song, W., Tan, K.H. and Baranek, A. (2009). Effective toolbox for lean service
implementation. International Journal of Services and Standards, 5 (1), 1-16
Windrum, P. and Tomlinson, M. (1999). Knowledge-intensive Services and International
Competitiveness: A Four Country Comparison. Technology Analysis and Strategic
Management, 11 (3), 391-408
Womack, J. P. and Jones, D. T. (1994). From Lean Production to the Lean Enterprise.
Harvard Business Review, 72, 93-103
Womack, J. P. and Jones, D. T. (1996). Beyond Toyota: How to Root Out Waste and
Pursue Perfection. Harvard Business Review, 74 (5), 140-158
Womack, James P., Daniel T. Jones, and Daniel Roos, The Machine that Changed the
222
World: Based on the Massachusetts Institute of Technology 5-Million-Dollar 5-Year
Study on the Future of the Automobile. Macmillan (1990).
Worley, J.M. and Doolen, T.L. (2006). The role of communication and management
support in a lean manufacturing implementation. Management Decision, 44 (2), 228-
245
Wu, L.C. and Wu, L.H. (2010). Service Engineering: An Interdisciplinary Framework.
Journal of Computer Information Systems, December (1), 14-23
Yasin, M.M., Small, M.H. and Wafa, M.A. (2003). Organizational modifications to
support JIT implementation in manufacturing and service operations. Omega, 31
(3),213-226
Yasuda, T. (2005). Firm Growth, Size, Age and behavior in Japanese manufacturing. Small
Business Economics, 24 (1), 1-15
Yavas, U. and Yasin, M.M. (2001). Enhancing organizational performance in banks: A
systematic approach. Journal of Services Marketing, 15 (6), 444 - 453
Yu-Lee, R.T. (2011). Proper Lean Accounting: Eliminating waste isn't enough; you have
to reduce inputs to save money. Industrial Engineer, 43 (10), 39-43
Zaman, K., Khan, M.M. and Ahmad, M. (2013). Factors affecting commercial energy
consumption in Pakistan: Progress in energy. Renewable and Sustainable Energy
Reviews, 19, 107–135
Zhang, D., Linderman, K. and Schroeder, R.G. (2012). The moderating role of contextual
factors on quality management practices. Journal of Operations Management, 30 (1-2),
12–23
Zu, X. (2009). Infrastructure and core quality management practices: how do they affect
quality? International Journal of Quality and Reliability Management, 26 (2), 129-149
223
Appendix (1): The questionnaire
Section I. Best Practices and Performance
1. Indicate the extent to which your firm has implemented the following practices: (tick one option) (Check the glossary
sheet for a definition of each expression if needed)
Practices
No
imlementation Considering Beginning Partially Substantially Fully
1 2 3 4 5 6
1- 5Ss
2- Automation
3- Continuous improvement
4- Group technology
5- Improving facility layout
6- Just in Time
7- Kaizen blitz
8- Kanban
9- Mistakes proofing/Poka-Yoke
10- Policy deployment/Hoshin Kanri
11- Process redesign
12- Pull system
13- Quality function deployment
14- Quick set up time
15- Root cause analysis
16- Single piece flow
17- Small lots
18- Standardisation
19- Takt time
20- Total preventive maintenance
21- Value stream mapping
22- Visualisation
23- Work load balancing
224
2. How much effort, in terms of monetary, human and other resources, did your firm extend on each of the activities listed below
as a direct consequence of implementing the practices reported in question (1)? (tick one option)
Activities
No
effort
Highest
level of
effort
1 2 3 4 5 6
1- An appropriate rewarding system
2- Effective Communication System
3- Employee empowerment
4- Employee commitment
5- Employee involvement
6- Having multifunctional employees
7- Leadership
8- Obtaining management support
9- Performance measurement system
10- Training
3. Indicate your level of agreement with achieving each of the listed benefits by your firm as a direct consequence of the
implementation of the practices reported in question (1): (tick one option)
Benefits Strongly
disagree
Strongly
agree
1 2 3 4 5 6
1- Freeing staff time
2- Identification and elimination of waste
3- Improvement in capacity
4- Improvement in customer perception of product/service quality
5- Improvement in customer satisfaction
6- Improvement in employees satisfaction and their performance
7- Improvement in employees understanding of the process
225
8- Improvement in operational efficiency
9- Improvement in productivity
10- Reduction in costs
11- Reduction in inventory
12- Reduction in lead time and cycle time
13- Reduction in the number of human errors
4. Has your firm formally implemented lean system? Yes No
If your answer is YES, indicate year initiated
Section II. Business strategy and product costing
5. Please indicate the level of your firm's emphasis on the following activities: (tick one option)
Activities No
emphasis
Highest
level of
emphasis
1 2 3 4 5 6
1- Achieving lower cost of services than
competitors
2- Making service/procedures more cost efficient
3- Providing high quality services
4- Customising services to customers need
5- Providing after-sale services and support
6- Introducing new services/procedures quickly
7- Improving the cost required for coordination of
various services
8- Improving the utilisation of available
equipment, services and facilities
226
9- Providing services that are distinct from that of
competitors
10- Offering a broader range of services than the
competitors
11- Improving the time it takes to provide services
to customers
6. Indicate by circling the costing system that is in use at your firm:
(1) Variable costing
(2) full absorption costing
(3) Activity-based costing
(4) Others
Section III. Demographics and Firm Characteristics
7. Approximately what is the percentage of foreign revenues to total revenues in this firm? (circle one option)
(1) (2) (3) (4) (5)
0% 1-25% 26-50% 51-75% 76-100%
8. Approximately what percent of employees at this firm is represented by a union(s)? (circle one option)
(1) (2) (3) (4) (5)
None 1–25% 26–50% 51–75% 76–100%
9. What is your job title?
10. How many years of experience do you have in managerial level? In management at your firm?
Thank you for your time and assistance in completing this survey. If you would like a copy of the results from this study, please check the box on the right.
E-mail address to receive a copy of the results:
227
Appendix (2): Introductory letter
Dear (participant name),
I am a doctoral student at Brunel University London whose PhD thesis focuses on the possible impact
of lean practices on the performance of UK service firms. This research is highly important given the
significant interest in lean concept among practitioners and academics and the noticeable lack of
research reporting on the effect of lean practices on the performance of service sector. Therefore, you
are kindly invited to voluntarily participate in this research, funded by myself, by answering the
enclosed questionnaire that will almost take 15-20 minutes of your time.
It is very important to note that your answers will be strictly confidential as approved by the Ethics
Committee of Brunel Business School. The aim of my research will be best achieved by reporting
information at an aggregate level (i.e. industry level), therefore the information you provide will not be
revealed at company level by any means. In addition to helping me completing my PhD programme,
your participation is highly important to improve the very limited knowledge available on service firms’
experience with lean system and therefore all participants are promised a copy of my results if they are
interested so that you can benchmark your practices against your industry.
This research seeks participation of randomly selected service firms operating in the UK including your
company, and you were specifically chosen due to being a member of your company current director
board believed to hold required knowledge to answer the questionnaire. However, please feel free to
share this questionnaire with other knowledgeable persons in your company for providing the most
accurate answers to the questions.
The questionnaire includes questions mainly focusing on your company’s experience with
implementing specific operations and management practices in its own operations. Therefore, if you are
going to participate in my research and that is my wish, please answer the questions from the
perspective of your current company ignoring any information related to subsidiaries owned by your
firm.
If you have any query related to your participation, please do not hesitate to contact me at
[email protected] and I will be happy to discuss about it.
Thank you very much for your help and cooperation.
Wael Hadid
PhD researcher
Brunel University
London.
228
Appendix (3): Glossary sheet
Practices Definition Reference
5Ss
Sort—sort out what is wanted in an area and what items can be disposed of,
reduced or moved, Set in order—place items to be retrieved closest to the area
for frequency of use and determine volume of use. Make visible so
abnormalities are apparent, Shine—make sure all items are in the best working
condition and remain so, Standardise—standardise work routines as well as
equipment and material usage, Sustain—ensure standards set are followed and
improved.
Esain et al.
(2008)
Automation It is the replacement of manual labour by advanced equipments. Bortolotti and
Romano (2010)
Continuous
improvement
A philosophy which promotes organisational change based on an ongoing
pattern of planning, execution and evaluation of results related to all
operations of an organisation for the purpose of forever improvement.
Emiliani (2004)
Group technology
Work processes are designed to form work cells which are located close to
each other with the object of cutting down on unneeded transport and waiting
times.
Suarez-Barraza
et al. (2009)
Changing the
facility layout
A layout designed according to optimum operational sequence or flow Suarez-Barraza
et al. (2009)
Just in Time It is the delivery of what is needed to where they are needed, in the quantity
needed, at the time they are requested.
Alagaraja
(2010)
Kaizen blitz Short-term process improvement projects that concern a specific area to
improve.
Suarez-Barraza
et al. (2009)
Kanban It is an information system that indicates when a subsequent activity within a
connected series of activities can start.
Manos et al.
(2006)
Mistakes
proofing/Poka-
Yoke
It is a process that helps eliminate the chance for mistakes. Manos et al.
(2006)
Policy
deployment/Hoshin
Kanri
A process used to connect corporate strategy to key objectives and resources,
including daily activities across functions.
Emiliani (2004)
229
Process redesign
To redesign content, scope, flow and structure of tasks and subtasks within an
organisation to enhance operational and customer-related performance
outcomes such as cost, productivity, quality, service, satisfaction and speed.
Yavas and
Yasin (2001)
Pull system To produce and deliver products/services at the request or pull of the customer
or user.
Manos et al.
(2006)
Quality function
deployment
Using a cross-functional team approach to reach consensus about final
product/service specifications, in accordance with customer requirements.
Alagaraja
(2010)
Quick set up time It is the ability to re setup an area for providing a different product/service
quickly.
Maguad (2007)
Root cause
analysis
Methods used to determine the root cause of a problem and identify
countermeasures to avoid repeat occurrences. Key tools are “5 Whys” (asking
why five or more times until the root cause of the problem is discovered) and
fishbone or cause-and effect diagram.
Emiliani (2004)
Single piece flow To pass the work to the next station right after finishing it without making any
batches.
Mirehei et al.
(2011)
Small lots To process transactions/information in the smaller batch possible and passed it
along to the next step.
Arbos (2002)
Standardisation It is an agreed-upon set of work procedures that establish the best and most
reliable methods and sequences for each process and each worker.
Kosuge et al.
(2010)
Takt time The rate of customer demand. Used to establish a direct link between
marketplace demand and workplace activities.
Emiliani (2004)
Total preventive
maintenance
A program used to ensure that equipment is in good operating condition and
available for use when needed.
Mirehei et al.
(2011)
Value stream
mapping
A visual picture of material and information flows from supplier to customer:
current-state map determines current conditions of flow; future-state map
shows opportunities for improvement at some future point.
Alagaraja
(2010)
Visualisation Signs and other forms of visual information used to simplify the workplace
and make it easy to recognise abnormalities.
Emiliani (2004)
Work load
balancing
It is the allocation of tasks in a balanced amount between employees so that
none will be over or under loaded with tasks.
Mirehei et al.
(2011)
